JP6942966B2 - Object detection device and mobile device - Google Patents

Description

The present invention relates to an object detection device and a mobile device.

In recent years, a device that projects light and receives light reflected by an object with a photodetector to detect information about the object such as the presence or absence of the object and the distance to the object has been actively developed.

For example, in the technique disclosed in Patent Document 1, when detecting information about a specific object (for example, a low reflection object), the threshold value for detecting the output signal of the photodetector is lowered.

However, in the apparatus disclosed in Patent Document 1, there is room for improvement in improving the detection accuracy of information regarding a specific object.

The present invention includes a light projecting system including a light source, a light receiving system including a light detector that receives light projected from the light projecting system and reflected by an object, and an output signal of the light detector or the output signal. A signal processing system that includes a signal detection unit to which a signal based on the light is input and outputs a detection signal, a first imaging system that captures at least a part of the projection range of the projection system and outputs a first imaging image, and the above. A second imaging system that captures at least a part of the projection range and outputs a second imaging image, and at least one region within the projection range of the projection system includes the detection signal , the first captured image, and the first image. 2 When the light projection condition of the light projection system and the processing condition of the signal processing system with respect to the attention area are projected to the attention area and the light projection is set as the area of interest based on the captured image. a control system for different between when projected in a region other than the region of interest in the region, an object detection apparatus Ru comprising a.

According to the present invention, it is possible to improve the detection accuracy of information about a specific object.

FIG. 5 is an external view of a vehicle equipped with a rider (non-scanning type) as an object detection device according to the first embodiment of the present invention. It is a block diagram for demonstrating the structure of the monitoring system of 1st Embodiment. It is a figure which shows the light-emitting waveform and the light-receiving waveform. It is a figure for demonstrating the light projection system and the light receiving system. It is a figure for demonstrating the relationship of a light source, a light receiving element, an optical system, and a detection area. It is a figure which shows the distance image generated by a rider. 7 (a) to 7 (c) are diagrams for explaining ch lighting patterns (No. 1 to No. 3) when a surface emitting laser array is used as a light source, respectively. 8 (a) and 8 (b) are diagrams showing a general relationship (No. 1 and No. 2) between the signal level and the noise level of the received signal, respectively. FIG. 9A is a diagram showing the relationship between the first signal level of the received light signal and the noise level, and FIG. 9B is a diagram showing the relationship between the second signal level of the received signal and the noise level. It is a flowchart for demonstrating the threshold value setting process. It is a flowchart for demonstrating the object detection process 1. It is a flowchart for demonstrating the SN ratio maintenance improvement process 1. It is a flowchart for demonstrating the SN ratio maintenance improvement process 2. It is a flowchart for demonstrating the SN ratio maintenance improvement process 3. It is a flowchart for demonstrating the object detection process 2. It is a flowchart for demonstrating the object detection process 3. It is a flowchart for demonstrating the object detection process 4. It is a flowchart for demonstrating the object detection process 5. It is a flowchart for demonstrating the object detection process 6. 20 (a) and 20 (b) are diagrams for explaining a lidar (scanning type) as the object detection device of the second embodiment, respectively. 21 (a) and 21 (b) are side views and front views of a vehicle equipped with an object detection device including the rider and the stereo camera of the third embodiment, respectively. It is a figure which shows the light projection range and the imaging range of the object detection apparatus of 3rd Embodiment. It is a figure which shows the object detection device which unitized a rider and a stereo camera. It is a figure which shows the appearance structure and the mounting example of the object detection apparatus of 4th Embodiment. It is a figure which shows an example of the hardware composition of the object detection apparatus of 4th Embodiment. It is a block diagram which shows an example of the software structure of the control device of the object detection device of 4th Embodiment. It is a block diagram which shows an example of the hardware composition of the control device of the object detection device of 4th Embodiment. It is a flowchart for demonstrating the object detection process 7.

Hereinafter, the first embodiment of the present invention will be described with reference to the drawings.

[First Embodiment]
First, an outline of the rider 20A as an example of the object detection device 100 of the first embodiment will be described.

<Overview>
FIG. 1 shows the appearance of a vehicle 1 (moving body device) in which a lidar (Light detecting and ranking) 20A is mounted on a vehicle body (moving body). The rider 20A is a distance measuring device that measures a distance by emitting laser light and detecting reflected light, and is also called a laser radar. The rider 20A of the first embodiment is a non-scanning type rider.

As an example, the rider 20A is attached near the license plate in front of the vehicle 1. The rider 20A may be arranged near the rear-view mirror of the vehicle 1, for example. The rider 20A receives power from, for example, the battery of vehicle 1.

In this specification, in the XYZ three-dimensional Cartesian coordinate system, the direction orthogonal to the road surface is described as the Z-axis direction, and the forward direction of the vehicle 1 is described as the + X direction.

FIG. 2 shows a block diagram of the configuration of the monitoring system 10 including the rider 20A and the monitoring control device 30. The monitoring control device 30 is installed in the vehicle 1. That is, the monitoring system 10 is mounted on the vehicle 1.

As an example, as shown in FIG. 2, the rider 20A receives a light receiving system 21 including an LD 21a as a light source and an LD driving unit 21b, a light receiving element 22a, an imaging optical system 22b, and a current / voltage converter 22c. The system 22 includes a signal processing system 24 including an averaging / integrating processing unit 24a and a binarization processing unit 24b, and a control system 23 for controlling the light source system 21 and the signal processing system 24. In addition, "LD" is an abbreviation for a laser diode.

In the monitoring system 10, the control system 23 of the rider 20A and the monitoring control device 30 are electrically connected. The monitoring control device 30 operates the rider 20A, and based on the information detected by the rider 20A, determines the shape and size of the object, calculates the position information of the object, calculates the movement information, and recognizes the type of the object. Etc. to determine the presence or absence of danger. When it is determined that there is a danger, it issues an alarm, turns the steering wheel to avoid the danger, and issues a command to step on the brake. These commands are sent to the alarm device, steering device, braking device, and the like of the vehicle 1, respectively.

<< Principle of object detection and ranging >>
In the first embodiment, the so-called direct TOF method is used as the basic principle of object detection and distance measurement (distance measurement) using the rider 20A. This direct TOF method will be briefly described with reference to FIG. 3, which is a conceptual diagram.

The control system 23 has a light projecting range depending on the presence or absence of a reflected light pulse (signal light) in the light receiving wave received by the light receiving system 22 when a light projecting wave including an emission pulse (signal light) is projected from the light projecting system 21. It is possible to determine the presence or absence of an object in.

Further, in the control system 23, from the timing when the LD 21a is pulsed (the timing when the light emitting pulse is emitted in the projected wave) to the light receiving timing of the light receiving element 22a (until the reflected light pulse reaches the light receiving element 22a in the light receiving wave). ) Is measured, and the measured value is multiplied by the speed of light to obtain the distance that the light reciprocates between the rider 20A and the object.

Since the light projecting system 21 and the light receiving system 22 are approximately the same distance from the object, and the distance from the LD 21a to the object and the distance from the object to the light receiving element 22a can be regarded as substantially the same, the control system 23 is as described above. Half of the obtained distance is calculated as the distance from the rider 20A to the object.

There are various methods for measuring time such as the optical time difference method and the optical phase difference method, and since there are many known examples, they are omitted here. Among the known examples, there is also a method called an indirect TOF method (see, for example, Japanese Patent Application Laid-Open No. 2015-501927, Japanese Patent Application Laid-Open No. 2013-538342, Japanese Patent No. 5089289). In the present invention, any method can be applied.

Here, in order to obtain the angular resolution, either the light projecting system 21 or the light receiving system 22 may be decomposed. When the light projecting system 21 decomposes, a plurality of light emitting areas that emit light independently are provided in the light source, and the plurality of light emitting areas are sequentially turned on. As a result, the angle range in which the light is projected can be determined, and the distance from the light emitting timing of each light emitting area to the object existing in the area corresponding to the light emitting area is measured by the time from the light receiving timing of the light receiving element to the light receiving timing of the light receiving element. can do.

In particular, when aiming at long-distance detection, it is difficult for a high-power end face emitting LD to have a plurality of emitting areas, so it is preferable to apply a surface emitting laser array, and as shown in FIG. 5, coupling. It is preferable to set a light projecting area in each area of the detection area (light projecting range) by using an optical system including a lens or the like.

When disassembling with the light receiving system 22, it is preferable to provide a plurality of light receiving areas in the light receiving element. In this case, even if the detection area is irradiated all at once, the light receiving element can independently detect the signal for each light receiving area, so that the detection area can be decomposed for each angle. In this case, as shown in FIG. 4, the region detected by each light receiving area is determined by using the imaging optical system 22b that collects the light emitted from the light source and reflected by the object into a plurality of light receiving areas of the light receiving element. It is preferable to set it.

In FIG. 5, each area of the light source or the light receiving element is shown only in a one-dimensional arrangement only in the Y-axis direction. It is also possible to impart angular resolution in the direction.

When the angular resolutions in the Y-axis direction and the Z-axis direction are obtained in this way, the distance information for each area of the detection region can be obtained, and the distance image of the detection region can be generated. That is, the detection area can be divided by the number of areas divided by the light source or the light receiving element to measure the distance, and one frame of distance image can be generated by one distance measurement for each divided area of the detection area. For example, when the light emitting area of the light source or the light receiving area of the light receiving element is divided into areas for each pixel, each divided area of the detection area corresponds to each pixel of the distance image on a one-to-one basis. Therefore, each divided area of the detection area is also referred to as a "pixel area".

Therefore, in the rider 20A of the first embodiment, a plurality of areas (light emitting area or light receiving area) of the light source or the light receiving element are two-dimensionally arranged along the YZ plane.

FIG. 6 shows the detection result (distance image) of the distance information to the object in a certain frame of the rider 20A.

In FIG. 6, of the three objects A, B, and C in the detection area, the object A is the closest to the rider 20A, and the object C is the farthest from the rider 20A. That is, the object A is the closest object.

Therefore, for example, when the distance from the rider 20A to the object A is less than a predetermined distance (for example, a predetermined distance of 100 m to 200 m), the non-attention mode for searching the entire projection range under uniform projection conditions, for example, the projection range. The mode shifts to the attention mode for detecting only the region where the object A, which is the closest object (the object closest to the rider 20A), exists in the above. When the mode shifts to the attention mode, the control system 23 sets the region where the object A exists as the attention region.

At this time, when a stationary object (an object that is constantly stationary) such as a road surface or a structure is detected at the same time, the control system 23 recognizes the stationary object and obtains the detection result excluding it. Based on this, it is determined which of the plurality of objects is the closest object.

Here, the region of interest is determined based on the frame shown in FIG. 6, but it is also possible to increase the number of frames to be referenced, including frames prior to this frame. For example, while recognizing the movement of a moving object around it over time, the flow line of an object within the detection range after the next frame can be predicted and reflected in the control of a moving object such as a vehicle 1, or it can move with a stationary object. It becomes possible to judge the distinction of an object.

When the control system 23 shifts to the attention mode, the control system 23 increases the amount of emitted light in the light emitting area of the light source, for example, in order to improve the SN ratio of the output signal (light receiving signal) of the light receiving system 22. The amount of emitted light can be increased by increasing the amplitude and duty of the drive current to the light emitting area (the former), and in the case of a light source that constitutes one light emitting area with a plurality of light emitting points, the light source is turned on. The amount of emitted light can also be increased by increasing the number of emission points (the latter). The drive current to the light emitting area can be controlled by adjusting the pulse amplitude and duty (pulse width / pulse period) of the modulated signal.

Here, the latter control will be described in detail with reference to FIG. FIG. 7 shows an example when the light emitting point (hereinafter referred to as “ch”) for projecting light into one area of the detection region is 36 ch. 7 (a) and 7 (b) are lighting patterns of the non-attention mode, respectively, and 18 channels different from each other are lit.

It is desirable to appropriately set the lighting pattern of the non-attention mode by setting the optical system to weakly divergent light so that the part of the ch (light emitting point) that is off is not missed. Further, from the viewpoint of the life of the ch, it is desirable to drive the ch so as to alternately switch between the lighting pattern of FIG. 7 (a) and the lighting pattern of FIG. 7 (b) even in the same non-attention mode.

The lighting pattern when the area is determined to be the area of interest from the non-attention mode is shown as an example in FIG. 7 (c). In this lighting pattern, all 36 channels are lit, and it is possible to increase the amount of projected light to the region of interest. An irradiation region (area irradiated from each channel) corresponding to each channel may be defined in advance, and the channels to be lit may be controlled so that the amount of projected light emitted to the region of interest increases.

Here, under the condition that the frame rates are the same, the light emission cycle (pulse cycle of the light emission pulse) of the light emission area that irradiates the region of interest can be shortened or the light emission duty (pulse width / pulse of the light emission pulse) can be shortened during one frame. By increasing the period), it is possible to increase the amount of projected light for each frame (the amount of light projected for each frame). The light emission cycle and light emission duty of the light emission area can be controlled by adjusting the pulse cycle and duty (pulse width / pulse cycle) of the modulation signal input to the LD drive unit 21b that drives the LD 21a.

It is also conceivable to reduce the frame rate only for the region of interest and increase the amount of projected light for each frame without changing the light emission cycle.

Then, the light receiving signal of the light receiving system 22 is monitored in the non-attention mode, and when the mode shifts to the attention mode, the object is repeatedly projected (irradiated) to the attention region until the signal level of the light receiving signal exceeds the threshold value. Information about an object (object information) such as presence / absence and distance to an object can be detected with high accuracy.

Further, the signal level of the received signal may be considered with respect to the noise level. For example, as shown in FIG. 8B, when the signal level does not protrude so much with respect to the noise level, light is continuously projected a plurality of times for each pixel region of the region of interest, and the light is received a plurality of times. By integrating and / or averaging the plurality of received signals of the light receiving system 22 by the averaging / integration processing unit 24a in the subsequent stage, the signal components are increased and the random noise is integrated and / or averaged in time. Since the noise level approaches zero, it is possible to improve the SN ratio of the obtained signal (signal based on the received signal). That is, as shown in FIG. 8A, the state is similar to the state in which the signal level of the received signal is sufficiently secured with respect to the noise level.

If the threshold value is lowered from the state shown in FIG. 8B as it is, only noise is detected and the object cannot be detected. Therefore, as shown in FIG. 8B, it is preferable to set the threshold value based on the maximum noise level assumed in advance, and in order to lower the threshold value below the set value, the number of times the received signal is integrated. It is necessary to reduce the noise level by increasing the number of noise levels and performing the averaging process. Alternatively, in the case of integration, the same effect can be obtained by raising or lowering the threshold value according to the number of integrations. Specifically, it is preferable to raise the threshold value as the number of times of integration increases and decrease the threshold value as the number of integrations decreases.

Further, in order to improve the SN ratio of the detection signal of the light receiving system 22, the noise level in the state where the light source is not turned on is held in advance, and the held noise level is subtracted from the received light waveform when the light source is turned on to subtract the reflected light. It can also be realized by extracting the waveform of the pulse (waveform of the signal light).

By performing the control as described above, sufficient signal light can be obtained from the region of interest with respect to noise and the like. Therefore, the distance measurement variation due to the noise of the object A existing in the region of interest is reduced, and the distance measurement variation up to the object A is reduced. Distance information can be detected with higher accuracy. Then, based on this distance information, deceleration / stop control (braking control) and avoidance control (steering control) of the moving body can be performed more precisely.

In the above, the case where an object is detected has been described, but when the object is not detected, the entire detection area is focused on in the non-attention mode in order to determine whether or not there is an undetectable object. It is preferable to consider it as an area and search with high accuracy.

As a result, it is difficult to detect unless the region of interest is set, for example, a target with low light reflectance such as black cloth, a large specular reflection component, and a small amount of light reflected / scattered to the rider 20A. It is possible to detect small objects such as mirrors and glass, which are small in size with respect to angular resolution and have a large loss of light intensity. These objects can be detected if they are close to the rider 20A to some extent, but the measures to be taken when they are detected at a close distance are limited to stopping the moving body. However, the deceleration of the moving speed is limited to the speed corresponding to the allowable stop time, and there is a concern of collision with an object. Therefore, by detecting an object before it approaches a close distance, it is possible to raise the upper limit of the moving speed of the moving body, and it is also possible to avoid traveling without reducing the moving speed.

Further, for example, as shown in FIG. 9A, the signal level of the first received signal, which is the received signal when the signal light from the object existing in the region of interest is received, is sufficiently higher than the noise level. At least one of the LD 21a and the averaging / integrating processing unit 24a may be controlled so as to exceed a large set threshold value 1. In this case, the signal-to-noise ratio of the signal level of the first received signal can be remarkably improved. Further, for example, as shown in FIG. 9B, the signal level of the second light receiving signal, which is the light receiving signal when the signal light from the object existing in the non-attention region is received, is larger than the noise level. At least one of the LD 21a and the averaging / integration processing unit 24a may be controlled so as to exceed the threshold value 2 set to be smaller than the threshold value 1 and fall below the threshold value 1.

As a result, the signal level of the first received signal can be made higher than the signal level of the second received signal. Further, the SN ratio of the first received light signal can be made larger than the SN ratio of the second received light signal.

The signal level of the first received signal may be controlled based on the threshold value 2 instead of the threshold value 1.

Hereinafter, the rider 20A of the first embodiment will be described in more detail.
<detail>
《Light projection system》
Returning to FIG. 2, as an example, the floodlight system 21 can independently drive each light emitting point with an LD21a (laser diode) having a light emitting point array composed of a plurality of light emitting points arranged two-dimensionally along a YZ plane. LD drive unit 21b and the like. A modulation signal (pulse signal) for each light emitting point is input from the control system 23 to the LD drive unit 21b. When the modulation signal for each light emitting point is input, the LD drive unit 21b supplies the driving current corresponding to the modulation signal to the light emitting point. At this time, an emission pulse is emitted from the emission point in the + X direction in front of the vehicle 1. The light emitted from the LD 21a is the light projected from the light projecting system 21. The light from each light emitting point travels while spreading at a predetermined diffusion angle. A region on the optical path of all the light from the LD21a and parallel to the YZ plane is referred to as a "detection region" or a "light projection range". The plurality of light emitting points correspond to a plurality of pixel regions of the detection region on a one-to-one basis.

Since the emission duty (emission time / emission cycle) at each emission point of the LD21a is limited from the viewpoint of the safety of the LD21a and the durability of the LD21a, it is desirable that the emission pulse has a narrow pulse width, and the pulse width is the same. , Generally, it is set to about several ns to several tens of ns. The pulse interval of the emission pulse is generally about several tens of microseconds.

Further, the size of the detection region can be adjusted by projecting the light from the LD21a through, for example, a lens. For example, if it is desired to widen the detection region, a concave power coupling lens may be provided on the optical path of the light from the LD21a, and if it is desired to be narrowed, a convex power coupling lens may be provided on the optical path of the light from the LD21a.

Further, when the light emitting region of the light emitting point of the LD21a cannot be ignored, a coupling state is set so that a conjugate image of the LD light emitting region appears at infinity according to the detection region, so that the light projection distribution can be easily made uniform. can. The light projection distribution can be controlled by using various optical element shapes such as a gable-shaped surface.

Further, when expanding the detection region, an optical element having a function of a diffuser such as a microlens array or frosted glass may be arranged on the optical path of the light from the LD21a.

Here, the light source is not limited to the LD, and various light sources such as a surface emitting laser (VCSEL) and an LED can be substituted.

《Light receiving system》
As an example, the light receiving system 22 includes a light receiving element 22a having a single light receiving area, an imaging optical system 22b, and a current-voltage converter 22c.

In FIG. 4, the light projecting system 21 and the light receiving system 22 are juxtaposed in the Y-axis direction, but they may be stacked in the Z-axis direction.

As the light receiving element, a CMOS imaging device (hereinafter, TOF sensor) having a PD (photodiode), an APD (avalanche photodiode), a Geiger mode APD SPAD (single photodiode avalanche diode), and a TOF (Time of Flight) calculation function for each pixel. ) Etc. Since APD and SPAD have high sensitivity to PD, they are advantageous in terms of detection accuracy and detection distance, and are desirable. Therefore, in the first embodiment, the APD is used as the light receiving element 22a.

Here, when there is an object in the detection area, the light beam from the light source is reflected or scattered by the object. The reflected light and scattered light are collected on the light receiving element 22a by the imaging optical system 22b. The imaging optical system 22b may have a configuration capable of condensing light on a lens system, a mirror system, or other light receiving element 22a. The output current of the light receiving element 22a is converted into a voltage signal (light receiving signal) by the current-voltage converter 22c, and the voltage signal is sent to the signal processing system 24. Here, a "photodetector" is configured including a light receiving element 22a and a current-voltage converter 22c. The photodetector may have a structure capable of detecting light, and may be, for example, a photodetector that detects heat generated by light incident and detects light.

By sequentially lighting a plurality of light emitting points of the LD21a and monitoring the presence or absence of a light receiving signal for each lighting, a light emitting point corresponding to the light receiving signal (light emitting point corresponding to the reflected light from an object) can be specified. Therefore, from the position of the specified light emitting point in the array, the pixel region in which the object exists in the detection region can be specified, and the position information of the object can be obtained.

<< Signal processing system >>
As an example, the signal processing system 24 includes an averaging / integrating processing unit 24a and a binarization processing unit 24b.

The averaging / integration processing unit 24a averages and / or integrates a plurality of light receiving signals sequentially output from the light receiving system 22, and outputs the obtained signal to the binarization processing unit 24b.

The binarization processing unit 24b binarizes (detects) the input signal when a threshold value for signal detection is set and the signal level of the input signal is higher than the threshold value, and controls the binarized signal as a detection signal. Output to system 23.

In the signal processing system 24, the averaging / integrating processing unit 24a can be omitted. That is, the light receiving signal may be directly output from the light receiving system 22 to the binarization processing unit 24b.

《Control system》
As an example, the control system 23 includes a measurement control unit 23a, a region of interest setting unit 23c, a time measurement unit 23d, and a distance calculation unit 23e.

Upon receiving the measurement start request from the monitoring control device 30, the measurement control unit 23a generates a modulation signal (pulse signal) and outputs the modulation signal to the LD drive unit 21b and the time measurement unit 23d. The measurement control unit 23a adjusts at least one of the pulse amplitude, pulse period, and pulse width of the modulated signal, if necessary.

Further, the measurement control unit 23a sets a threshold value for binarizing (detecting) the received light signal in the threshold value setting process described later, and outputs the setting information (threshold value setting information) to the binarization processing unit 24b. ..

Further, the measurement control unit 23a determines the presence / absence of an object in the detection region based on the presence / absence of the detection signal from the signal processing system 24, and when it is determined to be "presence", obtains the position information of the object based on the detection signal and is required. The attention area setting request and the position information of the object are sent to the attention area setting unit 23c according to the above.

Upon receiving the attention area setting request from the measurement control unit 23a and the position information of the object, the attention area setting unit 23c sets at least one area in the detection area where the object exists as the attention area based on the position information. The attention area setting information, which is the setting information, is output to the measurement control unit 23a.

The time measurement unit 23d obtains the light reception timing at the light receiving element 22a based on the detection signal (binarization signal) for each pixel region from the binarization processing unit 24b, and the light reception timing and the measurement control unit 23a The time difference from the rising timing of the modulated signal is obtained, and the time difference is output to the distance calculation unit 23e as the measurement time. As a method of obtaining the light receiving timing, for example, a timing between two timings at which the light receiving signal crosses the threshold value (for example, an intermediate timing) or a timing at which the light receiving signal crosses the threshold value from the bottom to the top can be considered.

The distance calculation unit 23e obtains the round-trip distance to the object by converting the measurement time for each pixel region from the time measurement unit 23d into a distance, and outputs 1/2 of the round-trip distance as distance data to the monitoring control device 30. Then, if necessary, it is also output to the measurement control unit 23a. The distance data for each pixel area from the distance calculation unit 23e constitutes the above-mentioned distance image as a whole.

The monitoring control device 30 performs, for example, steering control (for example, auto steering), speed control (for example, auto braking) of the vehicle 1 based on the distance data (distance image) from the distance calculation unit 23e.

The threshold value setting process performed by the measurement control unit 23a will be described below with reference to the flowchart of FIG. This threshold setting process is performed periodically (for example, every few minutes to several hours).

<< Threshold setting process >>
In the first step U1, the disturbance noise level is acquired. Specifically, the output of the light receiving element 22a when the LD21a is not emitting light is acquired as a noise level of disturbance noise (for example, noise due to disturbance light such as sunlight or illumination light, circuit noise, etc.).

In the next step U2, the acquired disturbance noise level is saved in the memory.

In the next step U3, a threshold value for binarizing the received signal is set. Specifically, a threshold value th larger than the disturbance noise level (more accurately, the maximum value of the disturbance noise level) is uniformly set for the entire projection range, and the setting information is sent to the binarization processing unit 24b. ..

Hereinafter, the object detection process 1 for detecting information about an object using the rider 20A of the first embodiment will be described.

<< Object detection process 1 >>
The object detection process 1 will be described with reference to FIG. The flowchart of FIG. 11 is based on a processing algorithm executed by the measurement control unit 23a. The object detection process 1 is started when a measurement start request is received from the monitoring control device 30. The monitoring control device 30 sends a measurement start request to the rider 20A, for example, when the electric system of the vehicle 1 on which the rider 20A is mounted is turned on.

In the first step S1, pulsed light is projected over the entire projection range. Specifically, a plurality of light emitting points of the LD21a of the light projecting system 21 are sequentially pulsed. That is, a modulation signal having the same pulse amplitude, pulse width, and pulse period between the light emitting points is applied to the LD drive unit 21b at different timings, and each light emitting point of the LD 21a is made to emit light at different timings with the same amount of light emitted.

In the next step S2, it is determined whether or not there is an object in the light projection range (detection area). Specifically, the presence / absence of the detection signal from the binarization processing unit 24b is monitored, and when the detection signal is "present", it is determined to be "with an object", and when it is "absent", it is determined to be "no object". If the judgment in step S2 is affirmed, the process proceeds to step S3, and if the judgment is denied, the same judgment is made again.

In step S3, the area where the object exists is specified. Specifically, a plurality of pixel regions corresponding to a plurality of light emitting points involved in the generation of the detection signal in step S2 are specified. That is, the position information of the object is specified.

In the next step S4, it is determined whether or not there are a plurality of objects. Specifically, it is determined whether or not there are a plurality of locations composed of the plurality of pixel regions specified in step S3. If the judgment here is denied, the process proceeds to step S5, and if affirmed, the process proceeds to step S8.

In step S5, the distance to the object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e based on the detection signal in step S2 is acquired.

In step S6, it is determined whether or not the distance to the object is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S7, and if denied, the process proceeds to step S13.

In step S7, the area where the object exists is set as the area of interest. When step S7 is executed, the process proceeds to step S11. Specifically, the attention area setting request and the position information of the object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the object surrounding the object as the attention area, and outputs the setting information to the measurement control unit 23a.

In step S8, the distance to each object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e based on the detection signal in step S2 is acquired.

In the next step S9, it is determined whether or not the distance to the closest object (the object closest to the rider 20A) is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S10, and if denied, the process proceeds to step S13.

In step S10, the region where the closest object exists is set as the region of interest. Specifically, the attention area setting request and the position information of the closest object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the closest object surrounding the closest object as the attention area (see FIG. 6). When step S10 is executed, the process proceeds to step S11.

In step S11, the “SN ratio maintenance / improvement process” is carried out. Details of the SN ratio maintenance / improvement process will be described later.

In step S12, the distance to each pixel region of the object of interest (the object existing in the region of interest) is calculated. The total of the calculated distances is the distance image of the object of interest. When step S12 is executed, the process proceeds to step S13.

In step S13, it is determined whether or not to end the process. The judgment here is affirmed when the measurement end request is received from the monitoring control device 30, and is denied when the measurement end request is not received. The monitoring control device 30 sends, for example, a measurement end request to the measurement control unit 23a when the electric system of the vehicle 1 on which the rider 20A is mounted is turned off. If the determination in step S13 is affirmed, the flow ends, and if denied, the process returns to step S2.

In the object detection process 1 described above, the position and size of the closest object are measured by focusing on the closest object in the range (range) of the rider 20A while increasing the SN ratio. , Shape and other information can be detected with high accuracy.

Next, "SN ratio maintenance / improvement process 1", "SN ratio maintenance / improvement process 2", and "SN ratio maintenance / improvement process 3", which are specific examples of the "SN ratio maintenance / improvement process" in step S11, will be described.

<< Maintenance and improvement process of SN ratio 1 >>
The SN ratio maintenance / improvement process 1 will be described below with reference to FIG. The flowchart of FIG. 12 is based on a processing algorithm executed by the measurement control unit 23a. Here, the signal processing system 24 does not have the averaging / integrating processing unit 24a.

In the first step T1, it is determined whether or not the received signal due to the reflected light from the object of interest is detected. Specifically, the presence / absence of a detection signal from the binarization processing unit 24b is monitored. In the binarization processing unit 24b, when a light receiving signal having a signal level higher than the threshold value th is input, the light receiving signal is binarized, and the binarized signal is output as a detection signal. If the judgment in step T1 is affirmed, the flow ends, and if it is denied, the process proceeds to step T2.

In step T2, the amount of projected light with respect to the region of interest is increased. Specifically, the pulse amplitude of the modulated signal corresponding to all the light emitting points corresponding to the region of interest is increased. Here, the amount of light projected on the region of interest is made larger than the amount of normal light projected, which is the amount of light projected in step S1 of the object detection process 1. When only a part of the light emitting points are made to emit light in step S1 of the object detection process 1, the number of light emitting points to be made to emit light in step T2 may be increased.

In the next step T3, it is determined whether or not the received signal due to the reflected light from the object of interest is detected. Specifically, the presence / absence of a detection signal from the binarization processing unit 24b is monitored. In the binarization processing unit 24b, when a light receiving signal having a signal level higher than the threshold value th is input, the light receiving signal is binarized, and the binarized signal is output as a detection signal. If the judgment in step T3 is affirmed, the process proceeds to step T4, and if denied, the process returns to step T2.

In this way, in steps T2 and T3, the amount of projected light to the region of interest is increased until the received signal due to the reflected light from the object of interest is detected. As a result, the SN ratio of the received signal can be improved. On the other hand, when the determination in step T1 is affirmed, the SN ratio of the received signal is maintained. When step T2 is performed a plurality of times, the amount of projected light increases as the steps T2 are repeated.

In the next step T4, the amount of projected light for the region of interest is returned to the normal amount of projected light. When step T4 is executed, the flow ends.

<< Maintenance and improvement process of SN ratio 2 >>
The SN ratio maintenance / improvement process 2 will be described below with reference to FIG. The flowchart of FIG. 13 is based on a processing algorithm executed by the measurement control unit 23a. Here, the signal processing system 24 has an averaging / integrating processing unit 24a.

In the first step T11, it is determined whether or not the received signal due to the reflected light from the object of interest is detected. Specifically, the presence / absence of a detection signal from the binarization processing unit 24b is monitored. In the binarization processing unit 24b, when a light receiving signal having a signal level higher than the threshold value th is input, the light receiving signal is binarized, and the binarized signal is output as a detection signal. If the judgment in step T11 is affirmed, the flow ends, and if it is denied, the process proceeds to step T12.

In step T12, the photoperiod with respect to the region of interest is shortened. Specifically, the pulse period of the modulated signal corresponding to all the light emitting points corresponding to the region of interest is shortened (the frequency is increased). Here, the photoperiod for the region of interest is projected once per frame in the normal projectile cycle, which is the projecting cycle in step S1 of the object detection process 1, but is projected a plurality of times per frame. The photoperiod is set short so that it is.

In the next step T13, the received signal due to the reflected light from the object of interest is averaged and / or integrated. Specifically, for each frame, a plurality of received light signals obtained by receiving a plurality of times by flooding a plurality of times are averaged and / or integrated. Thereby, the SN ratio of the received signal can be improved. On the other hand, when the determination in step T11 is affirmed, the SN ratio of the received signal is maintained. If a capacitor for accumulating the signal charge from the light receiving element 22a is provided, the signal charge of a plurality of light receiving signals can be accumulated in the capacitor for each frame, and the accumulated signal charge can be output at once.

In the next step T14, it is determined whether or not the received signal due to the reflected light from the object of interest is detected. Specifically, the presence / absence of a detection signal from the binarization processing unit 24b is monitored. In the binarization processing unit 24b, when a light receiving signal having a signal level higher than the threshold value th is input, the light receiving signal is binarized, and the binarized signal is output as a detection signal. If the judgment in step T14 is affirmed, the process proceeds to step T15, and if denied, the process returns to step T12.

In this way, the projection period for the region of interest is shortened until the received signal due to the reflected light from the object of interest is detected by steps T12 to T14. That is, the number of received signals that are averaged and / or integrated is increased for each frame. When step T12 is performed a plurality of times, the photoperiod cycle for each frame becomes shorter, the number of times of light reception increases, and the number of light-receiving signals averaged and / or integrated increases as the times are repeated.

In the next step T15, the photoperiod for the region of interest is returned to the original value (normal photoperiod). When step T15 is executed, the flow ends.

<< SN ratio maintenance and improvement processing 3 >>
The SN ratio maintenance / improvement process 3 will be described below with reference to FIG. The flowchart of FIG. 14 is based on a processing algorithm executed by the measurement control unit 23a. Here, the signal processing system 24 has an averaging / integrating processing unit 24a.

In the first step T21, it is determined whether or not the received signal due to the reflected light from the object of interest is detected. Specifically, the presence / absence of a detection signal from the binarization processing unit 24b is monitored. In the binarization processing unit 24b, when a light receiving signal having a signal level higher than the threshold value th is input, the light receiving signal is binarized, and the binarized signal is output as a detection signal. If the judgment in step T21 is affirmed, the flow ends, and if it is denied, the process proceeds to step T22.

In step T22, the frame rate with respect to the region of interest is lowered. Specifically, the frame rate for the region of interest is received once per frame at the normal frame rate, which is the frame rate in step S1 of the object detection process 1, but is received a plurality of times per frame in the region of interest. Decrease the frame rate so that it does.

In the next step T23, the received signal due to the reflected light from the object of interest is averaged and / or integrated. Specifically, for each frame, a plurality of received light signals obtained by receiving a plurality of times of light reception are averaged and / or integrated. Thereby, the SN ratio of the received signal can be improved. On the other hand, when the determination in step T21 is affirmed, the SN ratio of the received signal is maintained. If a capacitor for accumulating the electric charge from the light receiving element 22a is provided, the signal charge of a plurality of light receiving signals can be accumulated in the capacitor for each frame, and the accumulated signal charge can be output at once.

In the next step T24, it is determined whether or not the received signal due to the reflected light from the object of interest is detected. Specifically, the presence / absence of a detection signal from the binarization processing unit 24b is monitored. In the binarization processing unit 24b, when a light receiving signal having a signal level higher than the threshold value th is input, the light receiving signal is binarized, and the binarized signal is output as a detection signal. If the judgment in step T24 is affirmed, the process proceeds to step T25, and if denied, the process returns to step T22.

In this way, in steps T22 to T24, the frame rate with respect to the region of interest is lowered until the received signal due to the reflected light from the object of interest is detected. As a result, the SN ratio of the received signal can be improved. That is, the number of received signals that are averaged and / or integrated is increased for each frame. When step T22 is performed a plurality of times, the frame rate is lowered and the number of times of receiving light is increased, and the number of received light signals to be averaged and / or integrated is also increased.

In the next step T25, the frame rate for the region of interest is returned to the original value (normal frame rate). When step T25 is executed, the flow ends.

It should be noted that at least two of increasing the amount of light projected by the SN ratio maintenance and improvement process 1, shortening the projectile period of the SN ratio maintenance and improvement process 2, and lowering the frame rate of the SN ratio maintenance and improvement process 3. The SN ratio maintenance / improvement process 4 may be configured by combining the two.

Hereinafter, object detection processes 2 to 6 for detecting information about an object using the rider 20A of the first embodiment will be described.

<< Object detection process 2 >>
The object detection process 2 will be described with reference to FIG. The flowchart of FIG. 15 is based on a processing algorithm executed by the measurement control unit 23a. The object detection process 2 is started when a measurement start request is received from the monitoring control device 30. The monitoring control device 30 sends a measurement start request to the rider 20A, for example, when the electric system of the vehicle 1 on which the rider 20A is mounted is turned on.

In the first step S21, pulsed light is projected over the entire projection range. Specifically, a plurality of light emitting points of the LD of the light projecting system 21 are sequentially pulsed. That is, a modulation signal having the same pulse amplitude, pulse width, and pulse period between the light emitting points is applied to the LD drive unit 21b at different timings, and each light emitting point of the LD 21a is made to emit light at different timings with the same amount of light emitted.

In the next step S22, it is determined whether or not there is an object in the light projection range (detection area). Specifically, the presence or absence of the detection signal from the binarization processing unit 24b is monitored, and when the detection signal is "present", it is determined to be "with an object", and when the detection signal is "absent", it is determined to be "no object". do. If the judgment in step S22 is affirmed, the process proceeds to step S22.5, and if the judgment is denied, the same judgment is made again.

In step S22.5, the area where the object exists is specified. Specifically, a plurality of pixel regions corresponding to a plurality of light emitting points involved in the generation of the detection signal in step S22 are specified. That is, the position information of the object is specified.

In step S23, it is determined whether or not there is a moving object. For example, the change in the relative velocity between the object and the rider 20A is calculated from the position and frame rate of the object in a plurality of consecutive frames of the distance image, and when the change is equal to or more than the reference value, the object is determined to be a "moving object". If it is less than the reference value, the object is determined to be a "stationary object". If the determination in step S23 is denied, the process proceeds to step S24, and if affirmed, the process proceeds to step S27.

In step S24, the distance to the object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e is acquired based on the detection signal in step S22.

In the next step S25, it is determined whether or not the distance to the object is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S26, and if denied, the process proceeds to step S33.

In the next step S26, the region where the object exists is set as the region of interest. Specifically, the attention area setting request and the position information of the object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the object surrounding the object as the attention area, and outputs the setting information to the measurement control unit 23a. When step S26 is executed, the process proceeds to step S31.

In step S27, it is determined whether or not there are a plurality of moving objects. Specifically, it is determined whether or not the number of moving objects determined in step S23 is plural. If the determination here is denied, the process proceeds to step S24, and if affirmed, the process proceeds to step S28.

In step S28, the distance to each object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e is acquired based on the detection signal in step S22.

In the next step S29, it is determined whether or not the distance to the closest moving object (moving object closest to the rider 20A) is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S30, and if denied, the process proceeds to step S33.

In step S30, the region where the closest moving object exists is set as the region of interest. Specifically, the attention area setting request and the position information of the closest moving object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the closest moving object surrounding the closest moving object as the attention area (see FIG. 6).

In the next step S31, the “SN ratio maintenance / improvement process” is carried out. Here, for example, any of the above-mentioned SN ratio maintenance / improvement processes 1 to 3 is carried out.

In step S32, the distance to each pixel region of the object of interest (the object existing in the region of interest) is calculated. The total of the calculated distances is the distance image of the object of interest. When step S32 is executed, the process proceeds to step S33.

In step S33, it is determined whether or not to end the process. The judgment here is affirmed when the measurement end request is received from the monitoring control device 30, and is denied when the measurement end request is not received. The monitoring control device 30 sends, for example, a measurement end request to the measurement control unit 23a when the electric system of the vehicle 1 on which the rider 20A is mounted is turned off. If the determination in step S33 is affirmed, the flow ends, and if denied, the process returns to step S22.

In the object detection process 2 described above, the position, size, and shape of the closest moving object are measured by focusing on the closest moving object in the range of the rider 20A while increasing the SN ratio. Etc. can be detected with high accuracy.

<< Object detection process 3 >>
The object detection process 3 will be described with reference to FIG. The flowchart of FIG. 16 is based on a processing algorithm executed by the measurement control unit 23a. The object detection process 3 is started when a measurement start request is received from the monitoring control device 30. The monitoring control device 30 sends a measurement start request to the rider 20A, for example, when the electric system of the vehicle 1 on which the rider 20A is mounted is turned on.

In the first step S41, pulsed light is projected over the entire projection range. Specifically, a plurality of light emitting points of the LD of the light projecting system 21 are sequentially pulsed. That is, a modulation signal having the same pulse amplitude, pulse width, and pulse period between the light emitting points is applied to the LD drive unit 21b at different timings, and each light emitting point of the LD 21a is made to emit light at different timings with the same amount of light emitted.

In the next step S42, it is determined whether or not there is an object in the light projection range (detection area). Specifically, the presence or absence of the detection signal from the binarization processing unit 24b is monitored, and when the detection signal is "present", it is determined to be "with an object", and when the detection signal is "absent", it is determined to be "no object". do. If the judgment in step S42 is affirmed, the process proceeds to step S42.5, and if the judgment is denied, the same judgment is made again.

In step S42.5, the area where the object exists is specified. Specifically, a plurality of pixel regions corresponding to a plurality of light emitting points involved in the generation of the detection signal in step S42 are specified. That is, the position information of the object is specified.

In step S43, it is determined whether or not there is a moving object. Specifically, for example, the change in the relative velocity between the object and the rider 20A is calculated from the position and frame rate of the object in a plurality of consecutive frames of the distance image, and when the change is equal to or more than the reference value, the object is "moved". It is determined to be an "object", and if it is less than the reference value, the object is determined to be a "stationary object". If the determination in step S43 is denied, the process proceeds to step S45, and if affirmed, the process proceeds to step S44.

In step S45, the distance to the object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e based on the detection signal in step S42 is acquired.

In the next step S46, it is determined whether or not the distance to the object is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S47, and if denied, the process proceeds to step S53.

In the next step S47, the region where the object exists is set as the region of interest. Specifically, the attention area setting request and the position information of the object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the object surrounding the object as the attention area, and outputs the setting information to the measurement control unit 23a. When step S47 is executed, the process proceeds to step S51.

In step S44, it is determined whether or not there are a plurality of moving objects. Specifically, it is determined whether or not the number of moving objects determined in step S43 is plural. If the determination here is denied, the process proceeds to step S45, and if affirmed, the process proceeds to step S48.

In step S48, the distance to each object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e based on the detection signal in step S42 is acquired.

In the next step S49, it is determined whether or not there is a moving object whose distance is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S50, and if denied, the process proceeds to step S53.

In step S50, all regions in which moving objects having a distance less than a predetermined distance exist are set as regions of interest. Specifically, the attention area setting request and the position information of the moving object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the moving object surrounding the moving object whose distance is less than a predetermined distance as the attention region.

In the next step S51, the “SN ratio maintenance / improvement process” is carried out. Here, for example, any of the above-mentioned SN ratio maintenance / improvement processes 1 to 3 is carried out.

In step S52, the distance to each pixel region of the object of interest (the object existing in the region of interest) is calculated. The total of the calculated distances is the distance image of the object of interest. When step S52 is executed, the process proceeds to step S53.

In step S53, it is determined whether or not to end the process. The judgment here is affirmed when the measurement end request is received from the monitoring control device 30, and is denied when the measurement end request is not received. The monitoring control device 30 sends, for example, a measurement end request to the measurement control unit 23a when the electric system of the vehicle 1 on which the rider 20A is mounted is turned off. If the determination in step S53 is affirmed, the flow ends, and if denied, the process returns to step S42.

In the object detection process 3 described above, the position, size, shape, etc. of the moving object can be determined by focusing on the measurement while increasing the SN ratio for all the moving objects within the range of the rider 20A. Information can be detected accurately.

<< Object detection process 4 >>
The object detection process 4 will be described with reference to FIG. The flowchart of FIG. 17 is based on a processing algorithm executed by the measurement control unit 23a. The object detection process 4 is started when a measurement start request is received from the monitoring control device 30. The monitoring control device 30 sends a measurement start request to the rider 20A, for example, when the electric system of the vehicle 1 on which the rider 20A is mounted is turned on.

In the first step S61, pulsed light is projected over the entire projection range. Specifically, a plurality of light emitting points of the LD of the light projecting system 21 are sequentially pulsed. That is, a modulation signal having the same pulse amplitude, pulse width, and pulse period between the light emitting points is applied to the LD drive unit 21b at different timings, and each light emitting point of the LD 21a is made to emit light at different timings with the same amount of light emitted.

In the next step S62, it is determined whether or not there is an object in the light projection range (detection area). Specifically, the presence or absence of the detection signal from the binarization processing unit 24b is monitored, and when the detection signal is "present", it is determined to be "with an object", and when the detection signal is "absent", it is determined to be "no object". do. If the determination in step S62 is affirmed, the process proceeds to step S62.5, and if denied, the same determination is made again.

In step S62.5, the area where the object exists is specified. Specifically, a plurality of pixel regions corresponding to a plurality of light emitting points involved in the generation of the detection signal in step S62 are specified. That is, the position information of the object is specified.

In the next step S63, it is determined whether or not there is a moving object. For example, if the change in the relative velocity between the object and the rider 20A is greater than or equal to the reference value from the position and frame rate of the object in a plurality of consecutive frames of the distance image, the object is determined to be a "moving object" and is less than the reference value. In some cases, the object is determined to be a "stationary object". If the determination in step S63 is denied, the process proceeds to step S64, and if affirmed, the process proceeds to step S67.

In step S64, the distance to the object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e is acquired based on the detection signal in step S62.

In the next step S65, it is determined whether or not the distance to the object is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S66, and if denied, the process proceeds to step S74.

In step S66, the area where the object exists is set as the area of interest. Specifically, the attention area setting request and the position information of the object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the object surrounding the object as the attention area, and outputs the setting information to the measurement control unit 23a. When step S66 is executed, the process proceeds to step S71.

In step S67, it is determined whether or not there are a plurality of moving objects. Specifically, it is determined whether or not the number of moving objects determined in step S63 is plural. If the determination here is denied, the process proceeds to step S64, and if affirmed, the process proceeds to step S68.

In step S68, the distance to each object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e is acquired based on the detection signal in step S62.

In the next step S69, it is determined whether or not there is a moving object whose distance is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S70, and if denied, the process proceeds to step S74.

In step S70, the approach speed of all moving objects whose distance is less than a predetermined distance to the rider 20A is calculated. Specifically, the approach speed of the object is calculated from the change in the position of the object and the frame rate between consecutive frames of the distance image. The "approaching speed" is, for example, + when the moving object is approaching the rider 20A and-when it is separated.

In the next step S71, the region where the moving object having the fastest approach speed exists is set as the region of interest. Specifically, the attention area setting request and the position information of the moving object having the fastest approach speed are sent to the attention area setting unit 23c. At this time, the region of interest setting unit 23c sets a region of interest that is slightly larger than the moving object surrounding the moving object having the fastest approach speed.

In the next step S72, the “SN ratio maintenance / improvement process” is carried out. Here, for example, any of the above-mentioned SN ratio maintenance / improvement processes 1 to 3 is carried out.

In the next step S73, the distance of the object of interest (the object existing in the region of interest) to each pixel region is calculated. The total of the calculated distances is the distance image of the object of interest.

In step S74, it is determined whether or not to end the process. The judgment here is affirmed when the measurement end request is received from the monitoring control device 30, and is denied when the measurement end request is not received. The monitoring control device 30 sends, for example, a measurement end request to the measurement control unit 23a when the electric system of the vehicle 1 on which the rider 20A is mounted is turned off. If the determination in step S74 is affirmed, the flow ends, and if denied, the process returns to step S62.

In the object detection process 4 described above, the position, size, and position of the moving object are measured by focusing on the moving object having the fastest approaching speed in the range of the rider 20A while increasing the SN ratio. Information such as shape and moving speed can be detected with high accuracy.

<< Object detection process 5 >>
The object detection process 5 will be described with reference to FIG. The flowchart of FIG. 18 is based on a processing algorithm executed by the measurement control unit 23a. The object detection process 5 is started when a measurement start request is received from the monitoring control device 30. The monitoring control device 30 sends a measurement start request to the rider 20A, for example, when the electric system of the vehicle 1 on which the rider 20A is mounted is turned on.

In the first step S81, pulsed light is projected over the entire projection range. Specifically, a plurality of light emitting points of the LD of the light projecting system 21 are sequentially pulsed. That is, a modulation signal having the same pulse amplitude, pulse width, and pulse period between the light emitting points is applied to the LD drive unit 21b at different timings, and each light emitting point of the LD 21a is made to emit light at different timings with the same amount of light emitted.

In the next step S82, it is determined whether or not there is an object in the light projection range (detection area). Specifically, the presence or absence of the detection signal from the binarization processing unit 24b is monitored, and when the detection signal is "present", it is determined to be "with an object", and when the detection signal is "absent", it is determined to be "no object". do. If the determination in step S82 is affirmed, the process proceeds to step S82.5, and if denied, the same determination is made again.

In step S82.5, the area where the object exists is specified. Specifically, a plurality of pixel regions corresponding to the plurality of light emitting points involved in the generation of the detection signal in step S82 are specified. That is, the position information of the object is obtained.

In the next step S83, it is determined whether or not there is a low reflection object. Here, the signal level (light receiving amount) of the received light signal that is the source of the detection signal in step S82 is acquired, and the distance to the object obtained from the position information of the object obtained in step S82.5 of the signal level is relative to the distance to the object. When the ratio is less than a predetermined value, the object is regarded as a low reflection object. If the determination in step S83 is denied, the process proceeds to step S84, and if affirmed, the process proceeds to step S87.

In step S84, the distance to the object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e is acquired based on the detection signal in step S82.

In the next step S85, it is determined whether or not the distance to the object is less than a predetermined distance (for example, 100 m). If the judgment here is affirmed, the process proceeds to step S86, and if denied, the process proceeds to step S93.

In step S86, the area where the object exists is set as the area of interest. Specifically, the attention area setting request and the position information of the object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the object surrounding the object as the attention area, and outputs the setting information to the measurement control unit 23a. When step S86 is executed, the process proceeds to step S91.

In step S87, it is determined whether or not there are a plurality of low reflection objects. Specifically, it is determined whether or not there are a plurality of objects whose ratio obtained in step S82.5 is less than a predetermined value. If the determination here is denied, the process proceeds to step S84, and if affirmed, the process proceeds to step S88.

In step S88, the distance to each low-reflection object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e is acquired based on the detection signal in step S82.

In the next step S89, it is determined whether or not the distance to the closest low-reflection object (the low-reflection object closest to the rider 20A) is less than a predetermined distance (for example, 200 m). If the judgment here is affirmed, the process proceeds to step S90, and if denied, the process proceeds to step S93.

In step S90, the region where the closest low-reflection object exists is set as the region of interest. Specifically, the attention area setting request and the position information of the closest low reflection object are sent to the attention area setting unit 23c. At this time, the attention region setting unit 23c sets a region that is slightly larger than the closest low reflection object surrounding the closest low reflection object as the attention region.

In the next step S91, the "SN ratio maintenance / improvement process" is carried out. Here, for example, any of the above-mentioned SN ratio maintenance / improvement processes 1 to 3 is carried out.

In the next step S92, the distance of the object of interest (the object existing in the region of interest) to each pixel region is calculated. The total of the calculated distances is the distance image of the object of interest.

In the next step S93, it is determined whether or not to end the process. The judgment here is affirmed when the measurement end request is received from the monitoring control device 30, and is denied when the measurement end request is not received. The monitoring control device 30 sends, for example, a measurement end request to the measurement control unit 23a when the electric system of the vehicle 1 on which the rider 20A is mounted is turned off. If the determination in step S93 is affirmed, the flow ends, and if denied, the process returns to step S82.

In the object detection process 5 described above, the position and size of the closest low-reflection object are measured by focusing on the closest low-reflection object at a range of the rider 20A while increasing the SN ratio. Information such as shape can be detected with high accuracy.

<< Object detection process 6 >>
The object detection process 5 will be described with reference to FIG. The flowchart of FIG. 19 is based on a processing algorithm executed by the measurement control unit 23a. The object detection process 6 is started when a measurement start request is received from the monitoring control device 30. The monitoring control device 30 sends a measurement start request to the rider 20A, for example, when the electric system of the vehicle 1 on which the rider 20A is mounted is turned on.

In the first step S101, pulsed light is projected over the entire projection range. Specifically, a plurality of light emitting points of the LD of the light projecting system 21 are sequentially pulsed. That is, a modulation signal having the same pulse amplitude, pulse width, and pulse period between the light emitting points is applied to the LD drive unit 21b at different timings, and each light emitting point of the LD 21a is made to emit light at different timings with the same amount of light emitted.

In the next step S102, it is determined whether or not there is an object in the light projection range (detection area). Specifically, the presence or absence of the detection signal from the binarization processing unit 24b is monitored, and when the detection signal is "present", it is determined to be "with an object", and when the detection signal is "absent", it is determined to be "no object". do. If the determination in step S102 is affirmed, the process proceeds to step S102.5, and if the determination is denied, the same determination is made again.

In step S102.5, the area where the object exists is specified. Specifically, a plurality of pixel regions corresponding to a plurality of light emitting points involved in the generation of the detection signal in step S102 are specified. That is, the position information of the object is obtained.

In the next step S103, it is determined whether or not there is a low reflection object. Here, the signal level (light receiving amount) of the received light signal that is the source of the detection signal in step S102 is acquired, and the distance to the object obtained from the position information of the object obtained in step S102.5 of the signal level is relative to the distance to the object. When the ratio is less than a predetermined value, the object is regarded as a low reflection object. If the determination in step S103 is denied, the process proceeds to step S104, and if affirmed, the process proceeds to step S107.

In step S104, the distance to the object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e based on the detection signal in step S102 is acquired.

In the next step S105, it is determined whether or not the distance to the object is less than a predetermined distance (for example, 100 m). If the determination here is affirmed, the process proceeds to step S106, and if denied, the process proceeds to step S113.

In the next step S106, the region where the object exists is set as the region of interest. Specifically, the attention area setting request and the position information of the object are sent to the attention area setting unit 23c. At this time, the attention area setting unit 23c sets a region slightly larger than the object surrounding the object as the attention area, and outputs the setting information to the measurement control unit 23a. When step S106 is executed, the process proceeds to step S111.

In step S107, it is determined whether or not there are a plurality of low reflection objects. Specifically, it is determined whether or not there are a plurality of objects whose ratio obtained in step S103 is less than a predetermined value. If the determination here is denied, the process proceeds to step S104, and if the determination is affirmed, the process proceeds to step S108.

In step S108, the distance to each low-reflection object is acquired. Specifically, the distance data calculated by the distance calculation unit 22e based on the detection signal in step S102 is acquired.

In the next step S109, it is determined whether or not there is a low reflection object whose distance is less than a predetermined distance (for example, 100 m). If the determination here is affirmed, the process proceeds to step S110, and if denied, the process proceeds to step S113.

In step S110, all regions in which a low-reflection object having a distance of less than a predetermined distance exists are set as regions of interest. Specifically, the attention area setting request and the position information of the low reflection object are sent to the attention area setting unit 23c. At this time, the region of interest setting unit 23c sets a region that is slightly larger than the low-reflection object surrounding the low-reflection object as the region of interest.

In the next step S111, the "SN ratio maintenance / improvement process" is carried out. Here, for example, any of the above-mentioned SN ratio maintenance / improvement processes 1 to 3 is carried out.

In the next step S112, the distance of the object of interest (the object existing in the region of interest) to each pixel region is calculated. The total of the calculated distances is the distance image of the object of interest.

In the next step S113, it is determined whether or not to end the process. The judgment here is affirmed when the measurement end request is received from the monitoring control device 30, and is denied when the measurement end request is not received. The monitoring control device 30 sends, for example, a measurement end request to the measurement control unit 23a when the electric system of the vehicle 1 on which the rider 20A is mounted is turned off. If the determination in step S113 is affirmed, the flow ends, and if denied, the process returns to step S102.

In the object detection process 6 described above, the position, size, and shape of the low-reflection object are measured by focusing on all the low-reflection objects within the range of the rider 20A while increasing the SN ratio. Etc. can be detected with high accuracy.

In the object detection process 4, the object detection process 4'in which the "moving object" is replaced with the "low reflection object" may be performed.

Further, in each of the above object detection processes, when the determination in the second step (the step of determining whether or not there is an object) is denied, the "SN ratio maintenance / improvement process" is performed for the entire projection range. You may add a step to perform and then return to the second step after that step. In this case, it is possible to detect long-distance objects, low-reflection objects, small objects, etc., which are difficult to detect unless the SN ratio is improved.

Further, in each of the above object detection processes, it is preferable to perform detection by coarsening the non-attention region or lowering the frame rate by projecting light onto the non-attention region in parallel with a series of processes for setting the attention region. .. Then, when a specific object to be detected is detected in a certain non-attention region, it is preferable to switch the non-attention region to the attention region for detection.

[Second Embodiment]
20 (a) and 20 (b) show the configurations of the light emitting system and the light receiving system of the lidar 20B as an example of the object detection device 200 of the second embodiment.

The rider 20B is different from the rider 20A of the first embodiment in that the light projecting system has a scanning mirror that deflects and scans the light from the light source toward the effective scanning region. That is, the rider 20B of the second embodiment is a scanning type rider.

Here, a MEMS mirror for driving the mirror portion by a MEMS mechanism is used for this scanning mirror, but various modifications such as a polygon mirror in which a multifaceted mirror is rotated by a motor and other galvano mirrors are possible.

In the second embodiment, a photodetector (for example, a photodiode or a phototransistor) for detecting the light (scanning light) scanned by the MEMS mirror is provided in a light projection range (detection region) including an effective scanning region, and measurement control is performed. The unit detects the position of the scanning light (scanning position) based on the output of the light detection unit, and controls the light source of the projection system based on the scanning position. When the MEMS mirror is provided with a runout angle detection unit for detecting the runout angle of the mirror unit, the measurement control unit may detect the scanning position based on the output of the runout angle detection unit.

Further, the scanning configuration is such that the light emitting system and the light receiving system are overlapped in the Z-axis direction, and both are coaxially incident on the scanning mirror and scanned when viewed from the Z-axis direction as shown in FIG. As a result, the region in which the light is projected by the light projecting system and the region in which the light is taken in by the light receiving system are matched, and stable object detection is realized.

As a modification of the second embodiment, the scanning mirror scans only in the light projecting system, and the light receiving system observes the entire effective scanning region by the imaging optical system as in the first embodiment without using the scanning mirror. It can also be configured.

In this case, the space for installing the scanning mirror is not required in the light receiving system, and by reducing the size of the scanning mirror in the light projecting system, high-speed drive or wide-angle drive can be achieved.

In FIG. 20, the coordinate axes are set so that the effective scanning area is in the YZ plane, and the scanning mirror is not shown in FIG. 20, but can be scanned independently in the two axial directions of the Y-axis direction and the Z-axis direction. It is configured. When the scanning mirror is replaced with a polygon mirror or the like, it is also possible to arrange a plurality of reflecting surfaces at different angles with respect to the rotation axis and switch the scanning / detecting area in the Z-axis direction.

The method of irradiating the region of interest with light in the second embodiment will be described below.

Similar to the first embodiment, when the region where the object A exists is set as the region of interest based on the object detection result (distance image) as shown in FIG. 6, the drive of the MEMS mirror as the scanning means is set to the non-resonant mode. By switching, the mirror is driven only at an angle that deflects the light to the region of interest.

When it is desired to accurately detect the distance to the region of interest, the SN ratio can be increased by repeatedly scanning the region of interest to acquire a signal. When controlling a moving body so as to avoid the object A, it is necessary to detect the size and shape contour of the object A more accurately than the distance information to determine the movable area. Therefore, the MEMS mirror The drive of the object A may be further reduced so as to irradiate the contour portion of the object A, and the SN ratio may be improved at a higher speed.

In addition, without switching to the non-resonant mode, the light emission cycle of the light source when scanning the region of interest can be shortened, the emission duty can be increased, or the light source can be used while the mode remains in which the entire effective scanning region is scanned in the resonance mode. The SN ratio of the received signal may be increased by increasing the amplitude of the drive current to increase the peak light amount of the light emission pulse. This method is also effective in the case of an optical deflector that cannot be driven to deflect only a certain area, such as a polygon mirror.

If the light emission cycle and light emission duty of the light source cannot be changed due to restrictions on the driving conditions of the light source, the MEMS mirror may be set to the non-resonant mode as described above to irradiate the region of interest, or the scanning speed. It is also possible to substantially increase the number of times of light projection (the number of times of irradiation) to the region of interest by slowing down.

Using the rider 20B of the second embodiment described above, the above-mentioned threshold setting processing, SN ratio maintenance / improvement processing 1 to 3, and object detection processing 1 to 6 can also be performed.

[Third Embodiment]
In FIGS. 21 (a) and 21 (b), the vehicle 1'(mobile body device) in which the object detection device 300 including the rider 300a and the stereo camera 300b of the third embodiment is mounted on the vehicle body (moving body). A side view and a front view of the above are shown respectively.

The rider 300a has the same configuration as the rider 20A of the first embodiment or the rider 20B of the second embodiment, and is attached, for example, in the vicinity of the license plate on the front side of the vehicle 1'. The rider 300a may be mounted inside the vehicle 1', for example, but in that case, there is a concern that a part of the projected light from the rider 300a may be reflected by the windshield.

The stereo camera 300b is attached, for example, near the rear-view mirror in the vehicle 1', and detects the distance information of the detection area from the parallax images obtained from the imaging results of the imaging unit 1 (left eye) and the imaging unit 2 (right eye). do. The stereo camera 300b may be mounted outside the vehicle 1', but in that case, there is a concern that the lens of the stereo camera 300b may become dirty.

FIG. 22 shows the projection range of the lidar 300a and the imaging range of the stereo camera 300b.

As shown in FIG. 22, the imaging range of the imaging unit 1 (left eye imaging range) and the imaging range of the imaging unit 2 (right eye imaging range) have an overlapping region (overlapping portion), and the overlapping region thereof. The light projection range of the lidar 300a is included in. The overlapping region may match the light projection range or be included in the light projection range.

In the third embodiment, the imaging result of the stereo camera 300b is used to set the region of interest.

In general, a stereo camera is easier to increase the angular resolution than a rider and is excellent in detecting smaller objects, but it is difficult to increase the distance resolution as the distance increases, and the distance to an object can be detected accurately. Can not.

Therefore, after detecting the presence or absence of an object in the entire projection range using the stereo camera 300b and setting the attention region (after shifting to the attention mode), the distance to the object while increasing the SN ratio of the attention region by the rider 300a. By detecting the above, it is possible to detect information about a long-distance object or a low-reflection object with higher accuracy than in the non-attention mode.

Although the example of combining the rider and the stereo camera has been described in the third embodiment, the same effect can be obtained by combining the rider and the monocular camera. In the case of a monocular camera, for example, by taking an image through an optical filter, blurring or deviation is caused for each pixel, and distance information to the subject for each pixel is obtained depending on the degree of blurring or deviation. It is also possible to combine the rider with other sensing devices. Compared to lidar, millimeter-wave radar has excellent detection reliability in rain and fog, that is, weather resistance, but it is difficult to achieve both wide-angle detection area and improved angular resolution. For this reason, after roughly detecting an object in bad weather with a millimeter-wave radar to determine the region of interest, the lidar increases the SN ratio of the region of interest while detecting the object, which makes it more difficult in bad weather than in the non-attention mode. Highly accurate object detection can be realized.

Further, in the object detection device 300 of the third embodiment, the rider 300a and the stereo camera 300b are separately configured, but as shown in FIG. 23, the rider 300a and the stereo camera 300b are integrally configured. It may be installed near the license plate outside the vehicle or near the rearview mirror inside the vehicle.

Using the object detection device 300 of the third embodiment described above, it is also possible to perform the above-mentioned threshold setting processing, SN ratio maintenance / improvement processing 1 to 3, and object detection processing 1 to 6.

[Fourth Embodiment]
The object detection device 400 of the fourth embodiment including the integrally configured rider 300a and the stereo camera 300b shown in FIG. 23 will be described. FIG. 24 is a diagram showing an external configuration and an installation example of the object detection device 400.

As shown in the upper part (appearance configuration) of FIG. 24, the object detection device 400 includes a stereo camera 110 and a rider (laser radar) 120 as sensor devices (three-dimensional sensors) for acquiring the surrounding environment as three-dimensional information. And. The stereo camera 110 includes a monocular camera unit 111 (first imaging system) and a monocular camera unit 112 (second imaging system), and the rider 120 is arranged between the monocular camera unit 111 and the monocular camera unit 112. ..

The monocular camera units 111 and 112 each take a picture at a predetermined frame cycle while synchronizing with each other to generate a shot image.

The rider 120 measures the distance to the irradiation position (object) of the laser beam by irradiating the laser beam and receiving the reflected light.

As shown in the lower part (mounting example) of FIG. 24, the object detection device 400 is mounted, for example, at the center position inside the front window of the vehicle 140. At this time, both the stereo camera 110 and the rider 120 are attached toward the front of the vehicle 140. That is, in the vehicle 140, the object detection device 400 is attached so that the shooting direction of the stereo camera 110 and the emission direction of the laser beam of the rider 120 are in the same direction.

<< Hardware configuration of object detection device 400 >>
Next, an example of the hardware configuration of the object detection device 400 will be described with reference to FIG.

As shown in FIG. 25, the object detection device 400 has a camera stay 201 and a control board storage unit 202.

The monocular camera units 111 and 112 and the rider 120 are integrally attached to the camera stay 201. As a result, the object detection device 400 has been reduced in size and cost.

The control board storage unit 202 houses a control device (control system) including a signal processing system composed of a laser signal processing unit 240, a distance calculation processing unit 250, a memory 260, and an MPU (Micro Processing Unit) 270. ing. By configuring the laser signal processing unit 240 separately from the rider 120, the size of the rider 120 can be reduced. As a result, in the fourth embodiment, the rider 120 is arranged between the monocular camera unit 111 and the monocular camera unit 112.

In the example of FIG. 25, the laser signal processing unit 240 and the distance calculation processing unit 250 are configured as separate circuit boards, but the laser signal processing unit 240 and the distance calculation processing unit 250 have a common circuit. It may be composed of a substrate. This is because the cost can be reduced by reducing the number of circuit boards.

Subsequently, the details of each part on the camera stay 201 side will be described. As shown in FIG. 25, the monocular camera unit 111 (first image pickup system) includes a camera lens 211, an image pickup element 212, and a sensor substrate 213. The external light incident through the camera lens 211 is received by the image sensor 212 and photoelectrically converted at a predetermined frame period. The signal obtained by the photoelectric conversion is processed on the sensor substrate 213 to generate a captured image for each frame. The generated captured images are sequentially transmitted to the distance calculation processing unit 250 as comparative images.

The monocular camera unit 112 (second imaging system) also has the same configuration as the monocular camera unit 111, and the captured image generated in synchronization with the monocular camera unit 111 based on the synchronization control signal is a reference. As an image, it is sequentially transmitted to the distance calculation processing unit 250.

The rider 120 includes a light projection system including a light source drive circuit 231, a laser light source 232, and a light projection lens 233. The light source drive circuit 231 operates based on the synchronization control signal from the laser signal processing unit 240, and applies a modulation current (light source emission signal) to the laser light source 232. As a result, the laser beam is emitted from the laser light source 232. The laser light emitted from the laser light source 232 is emitted to the outside via the projection lens 233.

In the fourth embodiment, an infrared semiconductor laser diode (LD: Laser Diode) is used as the laser light source 232, and near-infrared light having a wavelength of 800 nm to 950 nm is emitted as the laser light. Further, the laser light source 232 periodically emits a laser beam having a pulsed waveform according to a modulation current (light source emission signal) applied by the light source drive circuit 231. Further, the laser light source 232 periodically emits a pulsed laser beam having a short pulse width of about several nanoseconds to several hundred nanoseconds.

The pulsed laser beam emitted from the laser light source 232 is emitted to the outside as a projection beam through the projection lens 233, and then is irradiated to a predetermined irradiation range. The irradiation range is preferably wider than the imaging range of the monocular camera unit 111 and the monocular camera unit 112. When expanding the irradiation range, it is necessary to increase the angular resolution of the light projecting system and the light receiving system. Therefore, by using a multi-channel surface emitting laser array as the laser light source or by laying out a large number of photodiodes in an array. It is preferable to use a light receiving element in which a large number of light receiving areas can be set.

The rider 120 further includes a light receiving system including a light receiving lens 234, a light receiving element 235, and a light receiving signal amplification circuit 236. The laser beam radiated to a predetermined object is scattered by the object. Then, the light component reflected along the same optical path as the laser light emitted from the rider 120 is guided to the light receiving element 235 as the reflected light via the light receiving lens 234.

In the fourth embodiment, a silicon PIN photodiode or an avalanche photodiode is used as the light receiving element 235. The light receiving element 235 generates a laser light receiving signal by photoelectrically converting the reflected light, and the light receiving signal amplification circuit 236 amplifies the generated laser light receiving signal and then transmits the generated laser light receiving signal to the laser signal processing unit 240.

Subsequently, the details of each part on the control board storage part 202 side will be described. The laser signal processing unit 240 is a signal processing system including an averaging / integrating processing unit and a binarization processing unit based on the laser light receiving signal transmitted from the rider 120. The laser signal processing unit 240 transmits the processed signal as a detection signal to the distance calculation processing unit 250.

The distance calculation processing unit 250 is composed of, for example, a dedicated integrated circuit such as an FPGA (Field-Programmable gate array) or an ASIC (Application Specific Integrated Circuit). The distance calculation processing unit 250 outputs a synchronization control signal for controlling the photographing timing and the laser light emission / reception timing to the monocular camera units 111 and 112 and the laser signal processing unit 240. The distance calculation processing unit 250 may be configured by an information processing platform composed of a CPU, ROM, RAM, and the like.

The distance calculation processing unit 250 generates a parallax image based on the comparison image transmitted from the monocular camera unit 111 and the reference image transmitted from the monocular camera unit 112. As a method for generating a parallax image, a known matching processing technique can be used, for example, SSD (Sum of Squared Difference), ZSD (Zero-mean Sum of Squared Difference), SAD (Sum of Absolute Difference), ZSAD. A matching method such as (Zero-mean Sum of Absolute Difference) can be used. Further, the distance calculation processing unit 250 sets a region of interest from the cluster information of the parallax image analyzed by the MPU 270 described later, transmits it to the laser signal processing unit 240 as a synchronization control signal, and transmits the laser light source and the laser signal processing unit 240. To control. Further, the distance calculation processing unit 250 generates distance information based on the detection signal transmitted from the laser signal processing unit 240. The distance calculation processing unit 250 stores the generated parallax image and distance information in the memory 260. The method of calculating the distance based on the detection signal from the laser signal processing unit 240 is the same as that of the first embodiment.

The memory 260 stores the parallax image and the distance information generated by the distance calculation processing unit 250. The memory 260 also provides a work area for the distance calculation processing unit 250 and the MPU 270 to execute various processes.

The MPU 270 controls each unit stored in the control board storage unit 202, and performs an analysis process for analyzing the parallax image and the distance information stored in the memory 260. One of the analysis processes includes a clustering process and an object detection process. The clustering process is, for example, a process of connecting pixels having similar parallax values in a parallax image as one cluster and dividing the parallax pixels into clusters. The parallax of each pixel of the parallax image may be replaced with a unit in the real space, and the parallax processing may be performed after the conversion processing as three-dimensional information. The cluster information calculated by the clustering process includes size information, position information, and color information (RGB information) of objects existing as each cluster. Further, the movement information of each cluster (information about movement including speed information and movement vector information) may be included in the cluster information by using it for a cluster of a plurality of frames. The MPU 270 also performs a clustering process on the distance image, and performs an object detection process using the cluster information calculated from the parallax image and the distance information. The clustering process for detecting an object from a parallax image or a distance image may be performed by a method other than the above, and for example, a clustering method such as a K-MEANS method, a CLARANS method, a BIRCH method, or a CURE method may be used.

<< Software configuration of control device >>
Next, the function of the control device (control system) of the object detection device 400 will be described. FIG. 26 is a functional block diagram showing an example of the function of the control device of the object detection device 400.

As shown in FIG. 26, the function of the control device includes a measurement control unit, a region of interest setting unit, a parallax image generation unit, an analysis processing unit, a time measurement unit, and a distance calculation unit. Each function is realized by the distance calculation processing unit (FPGA) 250, the memory 260, and the MPU 270. The measurement control unit is activated according to the control of the monitoring control device, receives signals from each of the first imaging system, the second imaging system, the floodlight system, the light receiving system, and the signal processing system, and controls each of them. Further, depending on the attention region set by the attention region setting unit, the detection accuracy improvement processing of the attention region (for example, for the attention region) is performed for at least one of the light receiving system, the light projection system, and the signal processing system. The amount of projected light is increased, the projecting period is shortened with respect to the region of interest, the number of times the received light signal is integrated with respect to the region of interest is increased, and the frame rate is decreased with respect to the region of interest). The parallax image generation unit generates a parallax image from the reference image and the comparison image acquired from the first imaging system and the second imaging system. The analysis processing unit performs clustering processing on the parallax image generated by the parallax image generation unit, and transmits the cluster information to the attention area setting unit. Further, the object detection process is performed from the distance information and the cluster information acquired from the signal processing system after the detection accuracy improvement process for the region of interest. The attention area setting unit sets the attention area based on the cluster information acquired from the analysis processing unit. The functions of the time measurement unit and the distance calculation unit are substantially the same as those described in the first embodiment. The fourth embodiment is different from the first embodiment in that the distance calculation unit sends the distance data only to the measurement control unit, and the measurement control unit sends the received distance data to the monitoring control device.

<< Hardware configuration of control device >>
Next, the hardware configuration of the control device (control system) of the object detection device 400 will be briefly described. FIG. 27 is a block diagram showing an example of the hardware configuration of the control device of the object detection device 400.

As shown in FIG. 27, the control device of the object detection device 400 is realized by a CPU, RAM, ROM, FPGA, and an external I / F. More specifically, the parallax image generation unit, the time measurement unit, and the distance calculation unit in FIG. 26 are realized by the FPGA, and the measurement control unit, the attention area setting unit, and the analysis processing unit in FIG. 26 are realized by the CPU.

<< Object detection process 7 >>
The object detection process 7, which is an example of the object detection process of the fourth embodiment, will be described with reference to FIG. 28.
The flowchart of FIG. 28 is based on a processing algorithm executed by the measurement control unit. The object detection process 7 is started when a measurement start request is received from the monitoring control device. The monitoring control device sends a measurement start request to the object detection device, for example, when the electric system of the vehicle equipped with the object detection device is turned on.

In the first step S201, the reference image is captured by the first imaging system, and the comparative image is captured by the second imaging system synchronized with the first imaging system.

In the next step S202, the parallax image generation unit performs matching processing on the reference image and the comparison image, calculates the parallax, and generates the parallax image.

In step S203, the analysis processing unit performs a clustering process. Specifically, the parallax in the parallax image is divided into clusters, cluster information including size information, position information, and movement information of each cluster is generated, and output to the attention area setting unit.

In step S204, the measurement control unit determines whether or not the cluster information output from the analysis processing unit exists. If the cluster information does not exist, the process proceeds to step S207, and if the cluster information exists, the process proceeds to step S205.

In step S205, the attention area setting unit sets the attention area based on the cluster information.

In step S206, the measurement control unit sets the rider for at least one of the light receiving system, the light projecting system, and the signal processing system so that the detection accuracy of the region of interest is improved. For example, settings such as increasing the amount of projected light for the region of interest, shortening the projecting cycle for the region of interest, increasing the number of times the received signal is integrated for the region of interest, and decreasing the frame rate for the region of interest. I do.

In step S207, distance measurement is performed by a rider (light projection system, light receiving system, signal processing system) according to the control of the measurement control unit, and distance information is acquired.

In step S208, the distance to the object existing in the detection area is measured from the cluster information and the distance information. If the distance to a predetermined object is different among the cluster information and the distance information, the distance information has priority.

In step S209, it is determined whether or not to end the process. The judgment here is affirmed when the measurement end request is received from the monitoring control device, and is denied when the measurement end request is not received.

By the above object detection process, the region of interest is set using the cluster information, and the distance measurement accuracy of the rider in the region of interest is improved, so that the cluster of interest can be measured with high accuracy. can. For example, in the cluster information, by measuring so that the distance measurement accuracy of the rider is high for the area where the closest cluster exists, the closest object area with a high possibility of collision can be measured. It is possible to perform high-precision rider distance measurement. In the fourth embodiment, the same non-scanning type rider as in the first embodiment has been described as an example, but the same scanning type rider as in the second embodiment may be adopted.

From the first viewpoint, the object detection devices 100, 200, 300, and 400 of the first to fourth embodiments described above are reflected by the light projecting system including the light source and the light projected from the light projecting system and reflected by the object. A light receiving system including a light detector having a light receiving element that receives the light, a signal processing system in which an output signal of the light detector or a signal based on the output signal is input, and a light projecting range within the light projecting range of the light projecting system. At least one region is set as the region of interest, and at least one of the light projection condition of the projection system and the processing condition of the signal processing system with respect to the region of interest is projected to the region of interest and the non-attention region (in the projection range). It is an object detection device provided with a control system that makes it different from when light is projected onto an area other than the area of interest.

In this case, for example, it is more advantageous for the control system to detect information about the object when at least one of the light projection condition and the processing condition is projected to the attention region than when the light is projected to the non-focus region. By setting the conditions, information about an object existing in the region of interest (a specific object) can be detected more accurately than information about an object existing in the non-region of interest.

As a result, the detection accuracy of information (object information) related to a specific object can be improved.

Further, from the second viewpoint, the object detection devices 100, 200, 300, and 400 of the first to fourth embodiments are projected by the light projecting system including the light source and the light projecting system and reflected by the object. A signal processing system including a light receiving system including a light detector having a light receiving element that receives light, and a binarization processing unit (signal detection unit) into which an output signal of the light detector or a signal based on the output signal is input. Then, at least one region within the light projection range of the light projection system is set as a region of interest, and the output signal of the light detector due to the projection of light into the region of interest or the signal based on the output signal is binarized by the binarization processing unit. It is an object detection device including a control system that controls at least one of a light projection system and a signal processing system so as to be binarized (detected).

In this case, the output signal of the photodetector (the signal based on the reflected light from the object existing in the region of interest) due to the projection of the light onto the region of interest can be detected with high accuracy.

As a result, the detection accuracy of information (object information) related to a specific object can be improved.

That is, according to the object detection devices 100, 200, 300, 400 of the first to fourth embodiments, regardless of the distance to the object, the presence or absence of movement of the object, the reflectance, size, shape, etc. of the object, It is possible to accurately detect information about an object such as the presence / absence of an object, the distance to the object, the position of the object, the size of the object, and the shape of the object. In particular, long-distance objects, low-reflection objects, small objects, etc. have a large loss of light quantity and have been difficult to detect in the past, but according to the object detection devices 100, 200, and 300, they can be reliably detected. Is possible.

In short, the object detection devices 100, 200, 300, and 400 of the first to fourth embodiments improve the SN ratio by systematic control after determining the region of interest, and can detect an object that has been difficult to detect until now. And. That is, it is possible to detect information about a longer distance, a lower reflection, and a smaller object.

Further, the control system may control the light projection system so that the amount of light projected on the region of interest is larger than the amount of light projected on the non-area of interest.

In this case, it is possible to increase the amount of reflected light from the region of interest and improve the SN ratio of the received signal.

Further, the control system may control the light projection system so that the light projection period for the attention region is shorter than the light projection period for the non-attention region.

In this case, the number of times of receiving the reflected light from the region of interest for each frame can be increased, and the SN ratio of the received signal can be improved.

In addition, the control system controls the light projection system to continuously project light to the region of interest a plurality of times, and the signal processing system averages a plurality of output signals of the photodetector due to the multiple light projections. It may further include an averaging / integrating processing unit that converts and / or integrates and outputs the obtained signal to the binarization processing unit. Specifically, the signal charge received by the photodetector and photoelectrically converted is stored in a capacitance (capacitor) to increase the number of times of storage when forming one frame. At this time, in the region of interest, the frame rate may be lowered or the emission frequency may be raised (the photoperiod may be shortened).

In this case, the signal light level can be improved and the SN ratio can be improved by continuously receiving the signal light (reflected light) from the object existing in the region of interest a plurality of times.

In addition, the threshold value th for signal detection is set in the binarization processing unit, and the control system acquires the disturbance noise level (noise level) from the output of the photodetector when the light is not projected from the light projection system. The threshold value th is set with reference to the disturbance noise level.

In this case, it is possible to control the signal level of the signal light (reflected light) from the object existing in the region of interest so as to be sufficiently higher than the disturbance noise level, and it is possible to improve the SN ratio. Further, by setting the threshold value to be larger than the disturbance noise level, erroneous detection can be prevented.

In order to determine the noise level, for example, the noise level, which is the received signal level when the light source is turned off, is held, and the noise level is subtracted from the received signal level when the light source is turned on. The latter level may be set as the signal light level, and the SN ratio may be increased by increasing the light source light amount, the light source period, and the charge accumulation time so that this level clears the reference value. Alternatively, even if the noise is not subtracted, if the noise level can be determined, the noise level and the signal light level can be compared, and the ratio, that is, the SN ratio is driven so as to exceed a predetermined reference value. It becomes possible to raise.

In addition, when there is no object in the projection range, the control system sets the projection conditions and signal processing conditions of the projection system for the entire projection range, and the projection conditions and signal processing of the projection system for the region of interest. It is preferable to make it equivalent to the processing conditions of the system. When an object is not detected in the entire detection area, that is, when there is no object nearby, it is not necessary to detect the object information in a hurry. , It is possible to detect smaller objects. As a result, it becomes possible to predict the surrounding environment from a longer-term perspective.

Further, the control system may project light only on the region of interest after setting the region of interest. In this case, the life of the light source can be extended and the power can be saved.

Further, the control system can set the region of interest based on the output of the signal processing system by the light projection from the light projection system. Specifically, the region of interest can be set by obtaining the presence / absence of an object, the position information of the object, and the movement information by using the detection result of the previous frame of the rider itself.

In this case, the region of interest can be set without separately providing another sensing device.

Further, in the object detection device 200 of the second embodiment, the projection system further includes scanning means for scanning the light from the light source, and the control system controls at least one of the light source and the scanning means.

As the scanning means, a MEMS mirror or a polygon mirror is assumed. In the case of the MEMS mirror, only the region of interest may be scanned, or if there is only one region of interest, the light beam may be stationary at an angle at which the light beam is deflected. In the case of a polygon mirror, by turning on the light source only when scanning the area of interest while continuing to rotate, the lighting time of the light source is suppressed, the life of the light source is extended, and the amount of light is increased only in the area of interest. It is possible to increase the SN ratio of the region of interest by increasing the accumulation time or the accumulation time.

Further, the scanning speed of the region of interest by the scanning means may be slower than the scanning speed of the region of interest, while the driving conditions of the light source are constant. In this case, the number of times the light beam is irradiated to the region of interest can be increased, the density of light intensity to the region of interest can be increased, and the SN ratio of the received signal can be improved.

Further, the light emission cycle of the light source with respect to the region of interest may be shortened while the operating conditions of the scanning means are constant. In this case, the number of times the light beam is irradiated to the region of interest can be increased, the density of light intensity to the region of interest can be increased, and the SN ratio of the received signal can be improved.

Further, the object detection device 300 of the third embodiment further includes a sensing device that senses the projection range, and the control system sets a region of interest based on the output of the sensing device.

In this case, by using a sensing device having characteristics different from that of the object detection device 300, it is possible to set the region of interest with higher accuracy.

Examples of the sensing device include a camera, a millimeter wave radar, and an infrared sensor. As for the camera, the angle resolution is higher and the detection of a smaller object is superior to that of the rider, but the distance resolution is low and the distance to the object cannot be detected accurately. Therefore, after the area of interest is determined more accurately by the camera, the object is detected while increasing the SN ratio of the area of interest by the rider, so that it is possible to realize object detection at a longer distance and with higher accuracy than in the non-attention state. .. Compared to lidar, millimeter-wave radar has excellent detection reliability in rain and fog, that is, weather resistance, but it is difficult to achieve both wide-angle detection area and improved angular resolution. For this reason, by performing rough object detection in bad weather with a millimeter-wave radar to determine the region of interest, and then detecting the object while increasing the SN ratio of the region of interest with the rider, it is possible to detect the object in worse weather than in the non-attention state. Highly accurate object detection can be realized. The infrared sensor is cheaper than the rider, but has a short detection distance and low accuracy. Therefore, it can be realized as a low-priced version of the above-mentioned system while suppressing the cost increase. Further, in addition to the detection result of the sensing device, the detection result before the previous frame of the rider itself may be taken into consideration to set the region of interest.

Further, the region of interest may be an region in which an object (an object within the range of the rider) whose distance from the object detection device is less than a predetermined distance (for example, 200 m) exists.

That is, an object outside the range of the rider may be excluded from the detection target because it is not necessary to detect it intensively.

It is possible to improve the detection accuracy of an object existing at a long distance by setting a long-distance region where the amount of light received tends to be weakened due to attenuation as a region of interest and improving the detection accuracy for the long-distance region. At this time, it is possible to determine whether or not the object is a long-distance object by using, for example, the position information included in the cluster information acquired by the stereo camera. Therefore, depending on the range of the rider (for example, when the range is 200 m or more), a region having a distance of 200 m or more from the rider may be set as the region of interest.

Further, the region of interest is preferably a region in which the object closest to the object detection device exists among the plurality of objects existing in the light projection range.

For example, in an autonomous moving body such as a robot or a safe driving support system (ADAS), the object that must be detected most urgently is the object closest to the moving body. Based on this result, in order to control the vehicle body to avoid or reduce the collision, it is necessary to acquire the three-dimensional information of the nearest object with higher accuracy. Therefore, by setting the region of interest to the region where the object closest to the object detection device exists, it is possible to accurately acquire the three-dimensional information of the nearest object and improve the avoidance accuracy by controlling the vehicle or the like. At this time, it is possible to determine whether or not the object is the closest object by using, for example, the position information included in the cluster information acquired by the stereo camera.

Further, the region of interest is preferably a region in which a moving object exists. This is because moving objects have lower predictability of relative position changes than stationary objects and have a higher risk of collision. At this time, it can be determined whether or not the object is a moving object by using, for example, the moving information included in the cluster information acquired by the stereo camera.

Further, the region of interest is preferably a region in which a moving object having the fastest approach speed to the object detection device exists among the regions in which a plurality of moving objects exist in the light projection range. This is because such a moving object has a particularly high risk of collision. At this time, it can be determined whether or not it is the fastest moving object by using, for example, the moving information included in the cluster information acquired by the stereo camera.

Further, the region of interest is preferably a region in which a low-reflection object exists. This is because low-reflection objects have a high risk of detection omission. At this time, it can be determined whether or not the object is a low-reflection object by using, for example, the color information included in the cluster information acquired by the stereo camera.

Further, according to the mobile device including the object detection devices 100, 200, 300, 400 of the first to fourth embodiments and the moving body on which the object detecting device is mounted, the movement having excellent collision safety is provided. Equipment can be provided.

The control system sets the first threshold value, which is a threshold value for detecting the output signal of the light detector due to the projection of light onto the region of interest, and the output signal of the optical detector due to the projection of light onto the region other than the region of interest. A second threshold value smaller than the first threshold value, which is a threshold value for detection, may be set. At this time, the control system needs to send information for identifying the light emitting unit involved in the projection to the region of interest and the light emitting unit involved in the projection to the non-attention region to the binarization processing unit. Specifically, a process of "raising the threshold value for the region of interest" may be added at the beginning of each SN ratio maintenance / improvement process.

In this case, the signal level of the signal light from the attention region can be surely higher than the signal level of the signal light from the non-attention region, and the SN ratio with respect to the attention region can be improved.

Further, the method of setting the region of interest in the object detection device is not limited to the method described in each of the above embodiments. In short, it is preferable to set a region of high priority that should be noted according to the speed of the moving body, the surrounding environment, etc. as the region of interest.

For example, when the moving speed of a moving object equipped with an object detection device is high (for example, when the vehicle is traveling at high speed), it is highly necessary to pay particular attention to the front, so the central region of the projection range is set as the region of interest. You may. Specifically, the light projection range is divided into three in a direction substantially perpendicular to the moving surface of the moving body such as the upper region, the central region, and the lower region, the central region is set as the region of interest, and the lidar with respect to the central region Increase the amount of light projected. As a result, it is possible to improve the accuracy of object detection in the central region, which is a region that needs to be watched when the moving speed is high.

For example, when the moving speed of the moving body is slow (for example, when the vehicle is traveling at low speed), it is highly necessary to pay attention to both sides (to deal with popping out, etc.), so the areas on both sides of the projection range are set as the areas of interest. You may. Specifically, the light projection range is divided into three in a direction substantially parallel to the moving surface of the moving body such as the left side region, the center region, and the right side region, and the left side region and the right side region are set as the attention region, and the left side region and the left side region are set. Increase the amount of light emitted by the rider to the right area. As a result, it is possible to improve the accuracy of object detection in the left side region and the right side region, which are regions that need to be watched when the moving speed is slow.

For example, when the object detection device has a lidar and a camera, the region of interest is set for the central region of the projection range using the image pickup result of the camera, and the lidar is detected for both regions of the projection range. The region of interest may be set using the result.

For example, when the object detection device has a lidar and a camera, the region of interest is set using the detection result of the lidar for the long-distance range of the floodlight range, and the camera is set for the short-distance range of the floodlight range. The region of interest may be set using the imaging result of.

For example, when the object detection device has a rider and a stereo camera, it is estimated that there is a moving object or a small object in a region where the parallax matching accuracy is low (for example, a region with a small amount of texture), and that region is set as a region of interest. You may. It can be estimated that the region with a large amount of texture is a road surface, a huge structure, or the like.

For example, when the object detection device has a rider and a monocular camera or a stereo camera, a region determined to have a small amount of texture from the luminance image of the camera may be set as a region of interest. In a stereo camera, the parallax matching accuracy is low in an area with a small amount of texture, so it is possible to improve the accuracy of object detection by focusing on the area using a rider whose accuracy is not so related to the amount of texture. can.

For example, when the object detection device has a rider and a stereo camera, a region determined to be an edge may be set as a region of interest. Specifically, the parallax image generated by the stereo camera is subjected to edge detection processing by, for example, a differential edge detection method, and a region determined to be an edge (a region in which similar parallax is continuous) is set as a region of interest. By detecting the edge with high accuracy in this way, it is possible to detect the existing region of the object with higher accuracy.

For example, for the area illuminated by the headlights of a vehicle equipped with an object detection device having a rider and a camera (bright area), the area of interest is set using the image pickup result of the camera, and for the dark area, the area of interest is set. The region of interest may be set using the detection result of the rider. This is because the rider has high detection accuracy even in a dark area. For bright areas, it is possible to detect with high accuracy even by using the image pickup result of the camera.
Further, a bright region (a region having a luminance value larger than a predetermined value) and a dark region (a region having a luminance value smaller than a predetermined value) may be detected using the image pickup result of the camera, and the dark region may be set as a region of interest. This is because a camera that can detect even low-reflection objects in bright areas has higher detection accuracy than a rider, so there is less need for intensive examination with a rider, but in principle it is a rider for dark areas. This is because the detection accuracy of the camera is higher than that of the camera, so it is necessary to intensively investigate with a rider. In this way, by intensively detecting the dark area as the area of interest with the rider, it is possible to improve the accuracy of object detection as compared with the case of the camera alone or the rider alone.

Further, in each of the above embodiments, an example in which the object detection device is mounted on a vehicle has been described, but the object detection device may be mounted on an unmanned aerial vehicle (for example, a drone) capable of remote control or autopilot, for example. In this case as well, by setting the region of interest, information on a specific object can be detected with high accuracy. At this time, in order to maintain the balance of the unmanned aerial vehicle in flight, an object detection device is provided at the bottom of the aircraft so that the center of gravity axis of the unmanned aerial vehicle and the center of gravity axis of the object detection device 300 are the same axis. The flight of the aircraft can be made more stable. Further, any of the object detection devices 100, 200, 300, and 400 of the first to fourth embodiments may be mounted on the unmanned aerial vehicle. In this case as well, by setting the region of interest, information on a specific object can be detected with high accuracy.

Further, the object detection devices 100, 200, 300, 400 of the first to fourth embodiments may be mounted on an autonomously movable robot. In this case as well, by setting the region of interest, information on a specific object can be detected with high accuracy, and by extension, the robot can be operated with high accuracy on the specific object.

That is, the object detection device of the present invention can be widely applied to the overall object information detection technology using TOF (time of flight).

Further, the object detection device of the present invention is not limited to the use mounted on a moving body, but can also be used for a use mounted on a stationary object or a device alone.

In addition, the numerical values, shapes, and the like used in the description of each of the above embodiments can be appropriately changed without departing from the spirit of the present invention.

The thinking process that led to the inventor's idea of each of the above embodiments will be described below.

As an object detection device, for example, in an in-vehicle application, a rider that detects the presence or absence of an object in front of a moving vehicle and the distance to the object is known. The rider can detect the presence or absence of an object and the distance to the object in a desired range by turning on the light source, projecting light, and detecting the light reflected or scattered from the object with a photodetector. So far, a combination of this with another distance measuring method such as a stereo camera has been disclosed.

In Patent Document 1 (Japanese Unexamined Patent Publication No. 2006-284293), the detection rate of a low-reflection object is improved by lowering the threshold value for detecting a signal in a region where a low-reflection object (low-reflection object) exists. ..

However, in Patent Document 1, the distance measuring performance is limited by the performance limit of each distance measuring device. Mainly, it is difficult to improve the SN ratio related to the detection distance and the accuracy from the limit performance of the distance measuring device.

In Patent Document 1, the threshold value is lowered when the object is a low-reflection object, but when the noise such as external light is large, the signal light becomes smaller than the noise in the first place, and the noise is detected even if the threshold value is lowered. However, the object cannot be detected. As a system, it is optimal to set the threshold value aiming at the maximum noise level assumed from the beginning, and it is not realistic to lower the threshold value.

Therefore, in order to solve the above problems, the inventor determines the state of interest in the object detection device, improves the SN ratio for the region of interest, and relates to a longer distance, lower reflection, and a smaller object. In order to detect information with high accuracy, we have come up with the idea of each of the above embodiments.

1, 1'... vehicle (mobile device), 21 ... light projecting system, 22a ... light receiving element (part of photodetector), 22c ... current / voltage converter (part of photodetector), 23 ... control system , 24 ... Signal processing system, 24a ... Average / integration processing unit (signal processing unit), 24b ... Binarization processing unit (signal detection unit), 100, 200, 300 ... Object detection device.

Japanese Unexamined Patent Publication No. 2006-284293

Claims (14)

A floodlight system including a light source and
A light receiving system including a photodetector that receives light projected from the light projecting system and reflected by an object, and a light receiving system.
A signal processing system that includes a signal detection unit to which an output signal of the photodetector or a signal based on the output signal is input and outputs a detection signal.
A first imaging system that captures at least a part of the projection range of the projection system and outputs a first captured image,
A second imaging system that captures at least a part of the light projection range and outputs a second captured image,
At least one region within the light projection range is set as a region of interest based on the detection signal, the first captured image, and the second captured image, and the light projection conditions of the light projection system and the signal processing system are set. at least one and a control system for different between when projected in a region other than the region of interest in the light projection range and when light is projected to the region of interest, the object detecting device Ru provided with processing conditions. The control system detects information about an object when at least one of the light projection condition and the processing condition is projected onto the region of interest than when the light is projected onto a region other than the region of interest. The object detection device according to claim 1, wherein the object detection device is set to favorable conditions. A floodlight system including a light source and
A light receiving system including a photodetector that receives light projected from the light projecting system and reflected by an object, and a light receiving system.
A signal processing system that includes a signal detection unit to which an output signal of the photodetector or a signal based on the output signal is input and outputs a detection signal.
A first imaging system that captures at least a part of the projection range of the projection system and outputs a first captured image,
A second imaging system that captures at least a part of the light projection range and outputs a second captured image,
At least one region within the projection range is set as a region of interest based on the detection signal, the first captured image, and the second captured image, and the output of the photodetector by projecting light onto the region of interest. a control system signal or a signal based on the output signal to control said at least one light projecting system and the signal processing system to be detected by the signal detection unit, an object detecting device Ru comprising a. The control system is characterized in that the light projection system is controlled so that the amount of light projected on the region of interest is larger than the amount of light projected on a region other than the region of interest in the projection range. The object detection device according to any one of 3. The control system is characterized in that the projection system is controlled so that the projection period for the region of interest is shorter than the projection period for a region other than the region of interest in the projection range. The object detection device according to any one of 4. The control system controls the light projection system to continuously project light on the region of interest a plurality of times.
The signal processing system further includes a signal processing unit that averages and / or integrates a plurality of output signals of the photodetector due to the plurality of projections and outputs the obtained signal to the signal detection unit. The object detection device according to any one of claims 1 to 5, wherein the object detection device is characterized. A threshold value for signal detection is set in the signal detection unit, and a threshold value for signal detection is set.
Claims 1 to 1, wherein the control system acquires a noise level from the output of the photodetector when no light is projected from the light projection system, and sets the threshold value based on the noise level. The object detection device according to any one of 6. When an object does not exist in the projection range, the control system applies the projection conditions of the projection system and the processing conditions of the signal processing system to the entire projection range, and the projection of the projection system to the region of interest. The object detection device according to any one of 1 to 7, wherein the light conditions and the processing conditions of the signal processing system are the same. The light source includes a plurality of light emitting units arranged in an array.
The object detection device according to any one of claims 1 to 8, wherein the control system individually controls the plurality of light emitting units. The object detection device according to any one of claims 1 to 8, wherein the photodetector includes a plurality of light receiving units arranged in an array. The floodlight system further includes scanning means for scanning the light from the light source.
The object detection device according to any one of claims 1 to 8, wherein the control system controls at least one of the light source and the scanning means. The object detection device according to any one of claims 1 to 11, wherein the control system generates a parallax image based on the first captured image and the second captured image. The object detection device according to any one of claims 1 to 12 , wherein the region imaged by the first imaging system and the region imaged by the second imaging system include the light projection range. The object detection device according to any one of claims 1 to 13.
A mobile device including a mobile body on which the object detection device is mounted.

Images