JP5577627B2 - Spectrum measuring device for moving objects - Google Patents

Description

  The present invention relates to a spectrum measuring apparatus for a mobile object that identifies a measurement object from spectrum data of the measurement object measured by a spectrum sensor mounted on a vehicle, particularly a mobile object such as an automobile.

  In recent years, vehicles such as automobiles are equipped with devices that support driving and decision making by recognizing the state of pedestrians and signals that change dynamically around the vehicle as driving assistance devices. Not a few. Many of such devices capture the state of signals, pedestrians, and the like with a CCD camera, etc., recognize the state by processing the captured image in real time, and recognize the result of the above-described driving. It is used for support. However, since the shape of the pedestrian usually changes variously depending on the size and orientation, the presence or absence of belongings, and the like, it is difficult to accurately recognize the presence from the shape obtained based on the image processing. In general, a traffic light has high standard for size and color, but it is difficult to avoid inconveniences such as change in shape depending on the viewing angle, and shape recognition through image processing is still limited.

  On the other hand, Patent Document 1 describes a remote sensing technique using spectrum data collected by a spectrum sensor as a technique for recognizing a measurement target. That is, here, from multispectral image data including invisible light areas captured by a spectrum sensor mounted on an aircraft or an artificial satellite, it is difficult to recognize only in the visible light areas such as forests, fields, and urban areas. The measurement object is classified and characterized, and the measurement object is identified based on the data thus classified and characterized.

JP 2000-255102 A JP 2006-145362 A

  As described above, in the spectrum sensor, since the luminance value (light intensity) of each wavelength band including the invisible light region is observed, it is possible to know the characteristic specific to the measurement object by comparing the luminance value for each wavelength. Can be identified. In recent years, a hyperspectral sensor having a wide imaging bandwidth and a high resolution of several nanometers to several tens of nanometers has been put to practical use as such a spectral sensor (see Patent Document 2).

  Therefore, recently, it has been studied to mount such a spectrum sensor in a vehicle such as an automobile and to identify various measurement objects around the vehicle using spectrum data captured by the spectrum sensor. However, since a moving object such as a vehicle is generally close to the measurement object, the relative distance to the measurement object greatly changes with the movement, and even between the respective spectrum data acquired at different times, Due to such a change in distance, the spectrum waveform (spectrum intensity and spectrum shape) changes. Then, due to such a change in the spectrum waveform, the identification accuracy of the measurement object naturally decreases.

The present invention has been made in view of such circumstances, and its purpose is a spectrum waveform resulting from movement of a moving body that occurs in imaging data (spectrum data) by a spectrum sensor mounted on a moving body such as a vehicle. It is an object of the present invention to provide a spectrum measuring apparatus for a moving body that can reduce the change of the measurement object and can more accurately identify a measurement object.

Hereinafter, means for solving the above-described problems and the effects thereof will be described.
According to the first aspect of the present invention, a spectrum sensor capable of measuring wavelength information and light intensity information is mounted on a moving body, and measurement around the moving body is performed based on a spectrum waveform of observation light detected by the spectrum sensor. A spectrum measuring apparatus for a mobile unit for identifying an object, wherein a dictionary waveform storing a spectrum waveform including wavelength information and light intensity information for a plurality of predetermined measuring objects as dictionary data, An arithmetic device that identifies the measurement object based on a comparison operation between the spectral waveform of the observation light and the spectral waveform stored in the dictionary data storage unit, and a moving state acquisition device that acquires information on the moving state of the moving body And the arithmetic unit is configured to transmit the observation light based on information on a moving state of the moving body acquired by the moving state acquisition device. And in the spectrum waveform to perform a comparison operation between the observed spectrum waveform stored corrected to the dictionary data storage unit to compensate for the attenuation due to atmospheric spectral intensity of which correlates with the distance between the measurement object, The gist is to make the correction value used for the correction different according to the presence or absence of the light .

  When acquisition of imaging data (spectral data) of a measurement object around the moving body, for example, a person or a traffic light, by a spectrum sensor mounted on a vehicle such as an automobile, for example, the moving body itself, that is, the moving body itself The relative distance from the object to be measured varies greatly depending on the movement state. For this reason, the spectral waveform (spectral intensity and spectral shape) changes between the spectral data acquired at different points in time due to the change in relative distance from the measurement object. As described above, it is possible to bring In this regard, as a moving body spectrum measuring apparatus that acquires information on the moving state of the moving body by the moving state acquisition apparatus and corrects the spectrum waveform of the observation light by the spectrum sensor based on the acquired information on the moving state. According to the above configuration, even if the moving state of the moving object, that is, the relative distance to the measurement object changes every time, for each attribute of the measurement object, that is, for each spectrum characteristic peculiar to a person or a traffic light. Each can be corrected to the same spectrum waveform. For this reason, even when the identification (tracking) of the measurement object is performed from the spectrum waveform based on the comparison with the dictionary data through the arithmetic unit, the identification accuracy is naturally improved.

In addition, as described above, the spectrum waveform (spectrum intensity) of the spectrum data acquired by the spectrum sensor changes according to the distance to the measurement target. However, the main factor is attenuation by the atmosphere, and environmental light. The spectrum intensity attenuates inversely proportional to the square of the distance under the passive condition where the light is not sunlight, and the distance is under the active condition where the ambient light is not the light lit from the moving body itself. The spectral intensity attenuates in inverse proportion to the fourth power of. Therefore, as described above, if the correction value used for the correction is made different depending on the presence or absence of the light lamp , the accuracy of the correction can be maintained at a high level.

The invention according to claim 8 is equipped with a spectrum sensor capable of measuring wavelength information and light intensity information in a moving body, and measures around the moving body based on a spectrum waveform of observation light detected by the spectrum sensor. A spectrum measuring apparatus for a mobile unit for identifying an object, wherein a dictionary waveform storing a spectrum waveform including wavelength information and light intensity information for a plurality of predetermined measuring objects as dictionary data, An arithmetic device that identifies the measurement object based on a comparison operation between the spectral waveform of the observation light and the spectral waveform stored in the dictionary data storage unit, and a moving state acquisition device that acquires information on the moving state of the moving body And the arithmetic unit is configured to transmit the observation light based on information on a moving state of the moving body acquired by the moving state acquisition device. A process of performing a comparison operation between the spectral waveform stored in the dictionary data storage section spectrum waveform is corrected to compensate for the attenuation due to atmospheric spectral intensity of the observation light correlates with the distance between the measurement object, a constant time in to the mobile body you move distance is short nearly as small fence, the amount of correction the distance is largely long nearly as, after correcting one of the spectral waveform of the observation light to the front and rear by the predetermined time, The gist of the present invention is to perform a process of determining the identity of the measurement object from which the observation light is acquired about the predetermined time based on the comparison between the corrected one and the other .

  Although it is somewhat lacking in accuracy, from the viewpoint of correction that compensates for the attenuation of the spectrum intensity due to the atmosphere of the observation light by the spectrum sensor, the distance from the measurement object is moved within a certain period of time. It can be approximated as a distance. That is, when the moving distance of the moving body within a certain time is short, the reflection spectrum from the measurement object has little effect on the change in the optical path length (the optical path length is not significantly different), so the correction amount is small. I'll do it. Conversely, when the moving distance of the moving body within a certain time is long, the reflection spectrum from the measurement object has a large influence on the change in the optical path length (a large difference occurs in the optical path length). Need to be bigger. Therefore, in this case, as described above, the correction amount with respect to the spectral intensity of the observation light is reduced as the approximate distance is shortened, and the correction amount with respect to the spectral intensity of the observation light is increased as the approximate distance is long. If the above correction is performed, practically useful correction can be performed.

According to a second aspect of the present invention, in the mobile body spectrum measuring apparatus according to the first aspect, the distance to the measurement object is a relative distance to the movement object, and the correction is performed on the measurement object or the movement object. The gist of the invention is that it is performed in such a manner that a constant spectral intensity is always obtained with respect to the change in relative distance with respect to the body.

If the distance to the measurement object can be obtained as a relative distance to the moving object, it is possible to correct more accurately the attenuation of the spectrum intensity caused by the atmosphere. In this case, as in the above configuration, if the correction of the spectrum intensity is performed in a manner in which a constant spectrum intensity is always obtained with respect to the change in the relative distance with respect to the measurement object or the moving body, It becomes possible to obtain a spectrum waveform that does not depend on a change in the distance to the measurement object. That is, here, when the moving object side is used as a reference, a larger correction (correction to reduce) is performed as the relative distance between the measuring object and the moving object becomes shorter, and conversely, when the measuring object side is used as a reference, As the relative distance between the measurement object and the moving object becomes longer, a larger correction (enlargement correction) is performed. Note that it is more desirable to use the mobile object as a reference if the measurement object is identified (tracked) from the spectrum waveform based on comparison with the dictionary data.

The invention according to claim 3, in the movable body spectrum measuring apparatus according to claim 2, distance information Information regarding the moving state of the moving body to obtain the moving state acquisition device for the measurement target of the mobile The gist is that the amount to be corrected is mapped in the correction data storage unit as a correction coefficient determined according to the acquired distance information each time.

  When the moving state acquisition device is a device that can acquire distance information from the moving body to the measurement target, as described above, the correction coefficient determined according to the acquired distance information as described above is used as the correction coefficient. It is desirable to map the amount to be corrected in the correction data storage unit. This makes it possible to obtain a correction coefficient corresponding to the distance information in each case through a simple map calculation. Examples of the movement state acquisition device that can acquire the distance information include a stereo camera and a radar device.

According to a fourth aspect of the present invention, in the spectrum measuring apparatus for a mobile unit according to the third aspect , the correction coefficient is a map having a plurality of values for each distance information for each of different wavelength bands of the spectrum waveform. The gist is to be realized.

  As described above, the attenuation of the spectrum intensity caused by the atmosphere cannot be ignored even by the influence of the atmospheric absorption band (wavelength band that is easily absorbed) according to the distance to the measurement target. Therefore, as in the above configuration, when mapping the correction coefficient, if the correction coefficient is mapped with a plurality of values for each speed information for each wavelength band having a different spectrum waveform, this is the case. It is possible to alleviate or avoid the effects of the atmospheric absorption band.

According to a fifth aspect of the present invention, in the mobile body spectrum measuring apparatus according to the third or fourth aspect, the presence or absence of light lighting of the mobile body as information on the mobile state of the mobile body acquired by the mobile state acquisition apparatus The gist of the present invention is that the correction coefficient is mapped to the correction data storage unit as another correction coefficient corresponding to the presence or absence of the light lamp.

  As described above, this is a light in which the ambient light is lit from the moving body itself, although the spectrum intensity attenuates in inverse proportion to the square of the distance under a passive condition where the ambient light is not sunlight. In other words, under active conditions, the spectral intensity attenuates in inverse proportion to the fourth power of the distance. Therefore, as described above, the spectral waveform correction factors described above are prepared separately for these passive conditions and active conditions, or for conditions that do not include an active state and conditions that include an active state. If this is done, it is possible to correct the influence of the ambient light.

According to a sixth aspect of the present invention, in the mobile spectrum measuring apparatus according to the fifth aspect of the present invention, the correction coefficient is mapped in such a manner that the correction for the presence of light lighting is larger than the correction for the absence of light lighting. It is the gist of what has been done.

  At least under active conditions, the spectrum intensity attenuates in inverse proportion to the fourth power of the distance. Therefore, as a specific correction coefficient, the correction for the light lamp is more effective than the correction for the no light lamp as described above. By mapping the correction coefficient in a manner that provides a large correction amount, the correction accuracy can be accurately maintained.

According to a seventh aspect of the present invention, in the mobile body spectrum measuring apparatus according to the third or fourth aspect, the correction coefficient is the correction data as a different correction coefficient corresponding to the spectrum waveform of the light irradiated by the light. The gist is that it is mapped in the storage unit.

  As described above, the spectrum intensity attenuates in inverse proportion to the fourth power of the distance under an active condition where the ambient light is not a light lit from the moving body itself. Therefore, as described above, if the above-described spectral waveform correction coefficients are prepared separately under active conditions, the influence of ambient light composed of such light lamps can be corrected. Become.

The block diagram which shows the outline of a structure of the apparatus as 1st Embodiment of the mobile body provided with the spectrum measuring apparatus for mobile bodies which concerns on this invention. The graph which shows the relationship between attenuation by an atmosphere, and the wavelength about the light intensity of the spectrum waveform of observation light. 6 is an example of a correction map in the embodiment, and is a map showing a relationship between a correction value for correcting the spectrum waveform of observation light and a relative distance between the measurement target. Explanatory drawing for demonstrating the relationship between the spectrum waveform of the measuring object in the same embodiment, and the distance of the vehicle with respect to a measuring object. It is a graph which shows the change of the spectrum waveform of observation light according to the change of the distance with the measuring object in the embodiment, (a) shows the case where the change of the distance is large, and (b) is the case where the change of the distance is small Indicates. The flowchart shown about the process of the spectrum analysis performed with the spectrum measuring apparatus in the embodiment. The map which shows the correction map used about 2nd Embodiment of the moving body provided with the spectrum measuring apparatus for moving bodies which concerns on this invention. FIG. 4 is an explanatory diagram for explaining an example of correcting a spectral waveform in the embodiment, where (a) is a diagram illustrating a state in which ambient light is passive, (b) is a diagram illustrating a state in which ambient light is active, (c) FIG. 4 is a diagram illustrating an example of a spectrum waveform of light, and FIG. 5D is a diagram illustrating an example of a corrected spectrum waveform. It is a state figure for explaining an example of relation with distance to a measuring object about a 3rd embodiment of a mobile object provided with a spectrum measuring device for mobile objects concerning the present invention, and (a) is a distance away. The figure which shows the time of being, (b) is a figure which shows when the distance is near. It is a graph which shows the change of the spectrum waveform of observation light according to the change of the distance with the measuring object in the embodiment, (a) shows the case where the change of the distance is large, and (b) is the case where the change of the distance is small Indicates. The flowchart which shows another aspect about the process of the spectrum analysis performed about a mobile body provided with the spectrum measuring apparatus for mobile bodies which concerns on this invention.

(First embodiment)
Hereinafter, a first embodiment in which a mobile object including the spectrum measuring apparatus for mobile object according to the present invention is embodied will be described with reference to the drawings.

FIG. 1 is a diagram showing a schematic configuration of the same device provided in a vehicle as a moving object equipped with a spectrum measuring device for moving object. As shown in FIG. 1, a vehicle 10 has a spectrum measurement device 11 that acquires optical information including visible light and invisible light outside the vehicle, and transmits information output from the spectrum measurement device to a passenger of a moving body. And a vehicle control device 13 that reflects the information output from the spectrum measurement device 11 in vehicle control. In addition, the vehicle 10 is provided with a vehicle state acquisition device 19 that can acquire information related to the movement state of the vehicle 10.

  The human machine interface 12 communicates a vehicle state and the like to a passenger, particularly a pilot, through light, color, sound, and the like, and an operation device such as a push button or a touch panel is provided so that the intention of the passenger is input through a button or the like. It is a known interface device provided.

  The vehicle control device 13 is one of the control devices mounted on the vehicle. For example, the vehicle control device 13 is directly connected to other various control devices mounted on the vehicle such as an engine control device or via an in-vehicle network. It is a device that can transmit information to each other. In this embodiment, the vehicle control device 13 transmits information such as an object identified by the spectrum measurement device 11 input to the device to other various control devices, and responds to the identified measurement object. So that the required driving assistance is executed on the vehicle 10.

  The spectrum measuring device 11 is provided with a spectrum sensor 14 that detects the spectrum data of the observation light, and a spectrum data processing device 15 that receives the spectrum data of the observation light detected by the spectrum sensor 14 and processes the data. . The spectrum sensor 14 separates observation light as light composed of visible light and invisible light into a predetermined wavelength band. Then, the wavelength information as information indicating each wavelength constituting the spectral wavelength band of the observation light, and the light intensity information as information indicating the light intensity of the same observation light dispersed at each wavelength for each wavelength. And output as spectral data. That is, the spectral data includes the spectral intensity that is the light intensity at each wavelength of the observation light, and the spectral shape formed from the distribution in the wavelength band of the same spectral intensity. A predetermined spectral waveform is obtained based on the spectral intensity and the spectral shape. The spectrum sensor 14 may measure the wavelength information and the light intensity information at the same time, or may measure each time when necessary.

  The spectrum data processing device 15 is mainly configured by a microcomputer having, for example, an arithmetic device and a storage device. The spectrum data processor 15 receives the spectrum data of the observation light detected by the spectrum sensor 14. The spectrum data processing device 15 identifies the measurement object observed based on the spectrum data of the input observation light and outputs the result, thereby outputting the result to the human machine interface 12 and the vehicle control device 13. .

  The spectrum data processing device 15 includes a dictionary data storage unit 16 in which the spectrum waveforms (spectrum data) of a plurality of measurement objects are stored as dictionary data, and the spectrum waveform of the measurement object as the dictionary data. An arithmetic unit 17 is provided that identifies a measurement object based on a comparison operation to be compared with a waveform. Further, the spectrum data processing device 15 is provided with a correction data storage unit 18 that holds correction data used for correcting the spectrum waveform of the observation light input to the arithmetic device 17.

  The dictionary data storage unit 16 includes all or part of a storage area provided in a known storage device, and spectrum data as dictionary data is stored in the storage area. Similar to the spectral data of the previous observation light, the spectral data as the dictionary data also includes wavelength information and light intensity information. From the spectral data, a predetermined spectrum is determined based on the spectral intensity and the spectral shape. A waveform is obtained.

The dictionary data is made up of spectrum data (spectrum waveform) of a measurement object as an object to be identified, and is prepared in advance for the number of measurement objects to be identified. Examples of measurement objects include pedestrians (people), bicycles, motorcycles, automobiles, and the like as moving objects, and non-moving objects include signals, signs, road surface paints, guardrails, stores, signboards, and the like. In addition, for example, pedestrians (people) are classified into children, elderly people, men, women, etc. according to more detailed attributes, and automobiles are classified into trucks, buses, sedans, SUVs, lighter according to more detailed attributes. You may classify into a car etc. That is, the storage area as the dictionary data storage unit 16 may be configured from storage areas of one or a plurality of storage devices so that a storage capacity capable of storing a plurality of dictionary data prepared in advance is satisfied.

  The data of one measurement object as dictionary data is provided as a pair with wavelength information corresponding to the number of light intensity information obtained by dividing the wavelength band that can be measured by the spectrum sensor by the wavelength resolution of the spectrum sensor. The amount of data is large. For example, when the wavelength band used for the comparison calculation is 400 to 2500 (nm) and the wavelength resolution is 5 (nm), one set of spectrum data includes 420 sets of wavelength information and light intensity information. It is included.

  In order to identify the measurement target, the calculation device 17 compares the calculation target dictionary data acquired from the dictionary data storage unit 16 with the spectrum waveform of the observation light, and obtains an identification result based on the identity and the like. Output. Further, the spectrum waveform of the observation light can be compared with the spectrum waveform of the observation light acquired in the past, and an identification result based on the identity can be output. Furthermore, various correction calculations including four arithmetic operations are performed on the spectrum waveform of the observation light, and the result can be obtained.

  The vehicle state acquisition device 19 is a variety of detection devices that acquire the state of the vehicle 10, in particular, the traveling state, and transmits the detected vehicle state to the connected computing device 17. As the vehicle state acquisition device 19, the vehicle speed that is the moving speed of the vehicle 10 is acquired from the speedometer, the acceleration of the vehicle 10 is acquired from the accelerometer, or the steering angle of the vehicle 10 is acquired as necessary. Each state of various equipment such as the lighting state of the light of the vehicle 10 may be acquired, or the distance to the measurement object measured by the distance sensor provided in the vehicle 10 may be acquired.

  By the way, the spectrum waveform of the observation light is affected by attenuation by the atmosphere. For this reason, when the spectral waveform of the observation light is compared with the spectral waveform of the dictionary data, the identification accuracy of the measurement object is improved by correcting the spectral waveform. In view of this, the correction data storage unit 18 holds correction values used when comparing the spectrum waveform of the observation light with the spectrum waveform of the dictionary data as map-like correction data (correction map). That is, in the present embodiment, the computing device 17 compares the spectrum waveform of the observation light corrected based on the correction value obtained from the correction map with the spectrum waveform of the dictionary data.

Next, correction of the spectral waveform of the observation light will be described with reference to FIGS.
FIG. 2 is a graph showing the relationship between the attenuation due to the atmosphere and the wavelength for the light intensity of the spectral waveform of the observation light. FIG. 3 is an example of a correction map, and is a diagram illustrating a relationship between a correction amount for correcting the spectrum waveform of the observation light and a relative distance between the measurement target as a map.

The spectral intensity of the observation light is changed depending on the wavelength band when it is attenuated by the atmosphere even if the emitted light from the measurement object has a uniform intensity in the entire wavelength band. That is, the atmosphere has an absorption band of light (a wavelength band that is easily absorbed), and the spectral waveform of the observation light is affected by the absorption band of the atmosphere. As shown in the graph La in FIG. 2, the atmospheric absorption band exists in a wavelength band near the wavelength fa and a wavelength band near the wavelength 2000 (nm), and has a light intensity in the same wavelength band as compared to other wavelength bands. Attenuation is large. For example, the wavelength fa as the absorption band of the atmosphere and the wavelength band formed in the vicinity thereof are wavelength bands having a high absorption rate by water, and the attenuation rate of the light intensity of the observation light is increased by moisture contained in the atmosphere.

  Further, the spectral intensity of the observation light is affected by the attenuation by the atmosphere, and the spectral intensity is lowered. At this time, the attenuation rate of light due to the atmosphere increases in a manner that is inversely proportional to the square of the measurement distance, which is the distance between the measurement object emitting the observation light and the spectrum sensor 14 that detects the observation light. That is, the correction amount when comparing the spectral waveform of the observation light acquired at a position away from the measurement object with the spectral waveform of the dictionary data is, for example, as shown in graphs Ca and Cb in FIG. Increases in proportion to the square of.

  This correction amount can also be used when comparing the spectral shapes of observation light detected at two different positions. In the case of the spectral shape of observation light detected at different positions, the difference in the distance between these detection positions is also the difference in the optical path length from the measurement target, so the spectral shape of one of the observation lights is used as the distance. By correcting with the correction amount corresponding to the difference in distance, it is possible to correct to at least the same intensity as the light intensity of the spectrum shape of the other observation light. For example, if there is one spectral shape that is close to the measurement target and has low attenuation and the other spectral shape that is far from the measurement target and has high attenuation, the other spectral shape is corrected to compensate for the attenuation caused by the difference in distance from the other. By applying, it is possible to obtain a light intensity suitable for comparison with the spectral shape of one measurement object. The graph Cb illustrates the correction rate when correcting a wavelength band including the wavelength fa corresponding to the atmospheric absorption band, and the graph Ca illustrates the correction rate when correcting other wavelength bands. .

  For these reasons, the correction amount for correcting the spectral waveform of the observation light is obtained based on the attenuation rate of the light intensity due to the wavelength band and the attenuation rate based on the distance to the measurement object. The correction amount obtained in this way is set in the correction map. Thus, by correcting the spectrum waveform of the observation light with the correction value set in the correction map, for example, a spectrum waveform suitable for comparison with the spectrum waveform of the dictionary data can be obtained.

  As a mode for comparing and calculating the spectral waveform of the observation light to be measured with the spectral waveform of the dictionary data, the distance corresponding to the spectral waveform of the dictionary data is used as a reference (for example, “0”), and correction is performed according to the difference from the distance. There is a mode in which the spectral waveform of the observation light is corrected by the value. Conversely, there is a mode in which the spectral waveform of the dictionary data is corrected with a correction value corresponding to the difference from the same distance. Which spectral waveform is to be corrected can be appropriately selected according to the purpose or the like, but for accurate identification, it is preferable to correct the spectral waveform of the observation light. By the way, when identifying an object to be measured, correction is not essential for the comparison operation. If the identity is determined by comparing characteristic portions of the spectrum waveform, such correction is not necessary. Even if it is not done, the measurement object can be identified from the spectral waveform of the observation light.

Further, as described above, a comparison operation of a plurality of, for example, two observation light spectra is also possible. That is, when comparing and calculating the spectrum waveform of two observation lights of the same measurement object acquired in a state where the distance to the measurement object is different, the spectrum waveform of the observation light separated from the measurement object is corrected to be close to the measurement object It can be compared with the spectrum of observation light. In addition, the spectral waveform of the observation light close to the measurement target is corrected and compared with the spectrum of the observation light at a distance, or the spectral waveform of each observation light is corrected, for example, with the spectral waveform of the dictionary data. You can also make comparisons. When identifying the measurement object with high accuracy, it is preferable to correct both spectral waveforms of the two observation lights and compare them with the spectral waveform of the dictionary data. However, when only determining the identity of the two observation light spectrum waveforms for tracking, etc., which is the process of recognizing a specific measurement object continuously in time, the spectrum waveform of one observation light is corrected. Then, it may be compared with the spectrum waveform of the other observation light. Even in this case, if at least one of the spectrum waveforms of the two observation lights is identified by comparison operation with the spectrum waveform of the dictionary data, the spectrum waveform of the other observation light is also indirectly However, the measurement target is identified by the comparison calculation with the spectrum waveform of the dictionary data.

  Next, the relationship between the spectrum waveform of the observation light and the vehicle distance will be described with reference to FIGS. 4 and 5. FIG. 4 shows the spectrum of observation light of “person” detected when the vehicle 10 equipped with the spectrum measuring device 11 approaches a person 31 as a measurement target that emits light based on sunlight received from the sun 30. It is explanatory drawing which shows a waveform. FIGS. 5A and 5B are graphs illustrating the comparison of the spectrum waveform of the observation light that changes with the distance to the measurement target, in which FIG. 5A shows a case where the change in distance is large, and FIG. 5B shows a small change in distance. Show the case.

  As shown in FIG. 4, at the time t = 0, the vehicle 10 detects the spectrum waveform L10 of the measurement object “person” separated by the separation distance Da. When the vehicle 10 approaches the person 31 at the speed S1, the traveling waveform Ma advances at the time t = 1, and the spectrum waveform L11 of the measurement object “person” that becomes the separation distance Db is detected. When the vehicle 10 further approaches the person 31 at the speed S2, the vehicle travels by a distance obtained by subtracting the movement distance Ma from the movement distance Mb at time t = 2, and the spectral waveform L12 of the measurement object “person” that becomes the separation distance Dc. Is detected. For convenience of explanation, it is assumed that the speeds S1 to S3 are the same speed.

  At this time, for example, if the speed S1 is twice the speed described above, the distance traveled per time is doubled, so the vehicle 10 travels by the moving distance Mb at time t = 1, and the separation distance Dc. The spectrum waveform L12 of the measurement target “person” is detected. That is, the detected spectrum shape is the same as the position of the vehicle 10 when the spectrum waveform of the observation light is acquired, more precisely “person”, even if the spectrum shape emitted by “person” is the same. It changes according to the distance to the vehicle. Needless to say, the position of the vehicle changes depending on the moving speed (vehicle speed) of the vehicle and the elapsed time. At this time, even if the person is moving, if the moving speed is relatively small, the moving distance Ma and the moving distance Mb can be approximated as distances approaching the person. In addition, when comparing the spectral waveforms of two observation lights, it is only necessary to make corrections based on the difference in distance between the detection positions, that is, the difference in the optical path length. It is also possible to apply the correction amount of the correction map by approximating the distance that the moving body moves. If the spectrum waveforms of the two observation lights are compared as described above, the correction value based on the moving distance of the vehicle 10 that can be easily acquired can be acquired even when the distance to the measurement target cannot always be obtained. Have.

  That is, as shown in FIG. 5A, when the spectrum waveform L10 detected when the time t = 0 is compared with the spectrum waveform L12 detected when the movement distance Mb is advanced from there, the spectrum waveform L12 is time t = The light intensity is higher by approaching the person 31 than the time of 0 by the moving distance Mb. In other words, the spectral waveform L10 has a low light intensity because it is separated from the person 31 by the moving distance Mb. When the distance increases, the light intensity decreases in inverse proportion to the square of the distance. Therefore, for example, the difference ΔD11 is substantially eliminated by correcting the spectrum waveform L10 with the correction value acquired from the correction map based on the movement distance. Thus, the shape can be approximated to the spectrum waveform L12.

As shown in FIG. 5B, when the spectrum waveform L10 detected when the time t = 0 is compared with the spectrum waveform L11 detected when the movement distance Ma is advanced therefrom, the spectrum waveform L11 is obtained at the time t = 0. By approaching the person 31 by the moving distance Ma, the light intensity is stronger. In other words, the spectral waveform L10 is weak in light intensity because it is separated from the person 31 by the moving distance Ma. When the distance increases, the light intensity weakens in inverse proportion to the square of the distance. Therefore, for example, the difference ΔD12 is substantially eliminated by correcting the spectrum waveform L10 with the correction value acquired from the correction map based on the moving distance. Thus, the shape can be approximated to the spectrum waveform L11.

  As described above, when the spectrum waveform L10 is approximated to the spectrum waveform L11 or the spectrum waveform L12 and the comparison operation is performed to determine the identity thereof, the tracking processing of the spectrum waveform L10 can be performed. Further, if the measurement target is identified by comparing the spectrum waveform L10 with the spectrum waveform of the dictionary data, the measurement target of the spectrum waveform L11 and the spectrum waveform L12 having the same identity as the spectrum waveform L10 can also be identified. Thus, the tracking process in which the measurement object is identified can be performed.

  Next, the spectrum analysis process in the spectrum measuring apparatus of this embodiment will be described with reference to FIG. FIG. 6 is a flowchart showing a spectrum analysis process including a tracking process performed by the spectrum measuring apparatus.

  This spectrum analysis process is a process repeatedly executed while the spectrum measuring apparatus 11 is activated. Here, for convenience of explanation, it is assumed that the spectrum analysis process for the spectrum shape at the time corresponding to time t = 0 in the above description has already been completed, and is the spectrum analysis process performed next. In addition, it is assumed that the measurement target is identified by the completed spectrum analysis process.

  When the spectrum analysis process is started, the spectrum data processing device 15 of the spectrum measuring device 11 detects the speed of the vehicle 10 and from the same speed and the elapsed time from the previous process, The distance moved is calculated (step S10 in FIG. 6). Further, the spectral data processing device 15 acquires the spectral waveform of the observation light that is input to the arithmetic device 17 as needed (step S11 in FIG. 6). When the spectrum waveform is acquired, a correction value corresponding to the previously calculated movement distance is selected from the correction value map, and the previous spectrum waveform is corrected by the selected correction value (step of FIG. 6). S12). Then, the identity between the corrected spectrum shape and the acquired spectrum shape is determined (step S13 in FIG. 6). In this embodiment, identity is performed based on a comparison operation that compares both spectral shapes, but the calculation method used to determine the identity is a variety of operations used in known image recognition techniques and data processing techniques. It may be a technique. Here, when the identity between the previous measurement object and the current measurement object is determined, the measurement object is tracked. In addition, since the measurement target of the previous spectrum shape is identified, the current spectrum shape is also identified by comparison with the dictionary data. As a result, the measurement target is identified, and the measurement target tracking process is performed.

  As a result, even if a change in the spectrum waveform of the observation light obtained from the spectrum sensor 14 due to the distance to the measurement object, that is, the moving distance of the vehicle occurs, the influence of such a change is suppressed and the identification accuracy is maintained high. To be able to.

As described above, according to the movable body spectrum measuring apparatus of the present embodiment, the effects listed below can be obtained.
(1) The configuration as the spectrum measuring device 11 is acquired by the vehicle state acquisition device 19 as the moving state of the vehicle 10 and information on the speed is acquired. Based on the acquired information on the moving state, the observation light from the spectrum sensor 14 is obtained. The spectrum waveform was corrected. Thereby, even if the moving state of the vehicle 10, that is, the relative distance to the measurement target changes every time, the same for each attribute of the measurement target, that is, for each spectrum characteristic peculiar to a person or a traffic light. It is possible to correct to a spectral waveform of

  (2) For this reason, even when the measurement object is identified (tracked) from the spectrum waveform based on the comparison with the dictionary data through the arithmetic unit 17, the identification accuracy is naturally improved.

  (3) As described above, the spectrum waveform (spectrum intensity) of the spectrum data acquired by the spectrum sensor 14 changes according to the distance to the measurement object, but the main factor is attenuation by the atmosphere. If the ambient light is not sunlight, the spectrum intensity attenuates in inverse proportion to the square of the distance under so-called passive conditions. Therefore, as described above, if the correction of the spectrum waveform described above is performed to compensate for the attenuation of the spectrum intensity correlated with the distance between the vehicle 10 and the measurement target due to the atmosphere, the accuracy of the correction is maintained high. Will come to be.

  (4) Although the accuracy is somewhat lacking, from the viewpoint of correction that compensates for the attenuation of the spectrum intensity due to the atmosphere of the observation light by the spectrum sensor, the distance from the object to be measured is calculated within a certain time. It can also be approximated as a moving distance. That is, when the distance that the vehicle 10 moves within a certain time is short, the reflection spectrum from the measurement object has little influence on the change in the optical path length (the optical path length is not significantly different), and the correction amount is small. I'll do it. Conversely, when the distance that the vehicle 10 moves within a certain time is long, the reflection spectrum from the measurement object has a large influence on the change in the optical path length (a large difference occurs in the optical path length). Need to be bigger. Therefore, in this case, as described above, the correction amount with respect to the spectral intensity of the observation light is reduced as the approximate distance is shortened, and the correction amount with respect to the spectral intensity of the observation light is increased as the approximate distance is long. If the above correction is performed, practically useful correction can be performed.

  (5) The distance traveled by the vehicle 10 can be converted into speed information by a speedometer (speed sensor) that is normally provided in the moving body itself. That is, if the speed of the vehicle 10 is low, the distance that the vehicle 10 moves within a certain time is short, and if the speed of the vehicle 10 is high, the distance that the moving body moves within a certain time is long. Therefore, as described above, if the correction amount is mapped to the correction data storage unit 18 as a correction coefficient determined according to the speed information thus converted, the sensors other than the spectrum sensor 14 are separately provided. Without correction, the above-described correction for the spectral intensity of the observation light can be performed through simple map calculation.

  (6) The attenuation of the spectral intensity caused by the atmosphere cannot be ignored due to the influence of the atmospheric absorption band (wavelength band that is easily absorbed) according to the distance to the measurement target. Therefore, as in the above configuration, when mapping correction coefficients, it is assumed that such correction coefficients are mapped with a plurality of values for each speed information for each wavelength band having different spectral waveforms. It is possible to reduce or avoid the influence of the absorption band.

  (7) The attenuation of the spectral intensity due to the atmosphere cannot be ignored due to the influence of the atmospheric absorption band (wavelength band that is easily absorbed) according to the distance to the measurement target. Therefore, as in the above configuration, when mapping the correction coefficient, if the correction coefficient is mapped with a plurality of values for each speed information for each wavelength band having a different spectrum waveform, this is the case. It is possible to alleviate or avoid the effects of the atmospheric absorption band.

(Second Embodiment)
Next, a second embodiment of the spectrum measuring apparatus for a moving body according to the present invention will be described with reference to FIGS. The present embodiment is the same in configuration as the first embodiment, and the correction map as correction data held in the correction data storage unit 18 is different. Hereinafter, a specific configuration as the spectrum measuring apparatus 11 will be mainly described focusing on differences from the first embodiment. Further, in FIG. 8, the same members as those in the first embodiment shown above are denoted by the same reference numerals, and redundant description is omitted.

  In the present embodiment, the distance between the spectrum sensor 14 and the “person” as the measurement target is measured by the vehicle state acquisition device 19, and the measured distance is input to the arithmetic device 17. To do. In addition, the vehicle state acquisition device 19 acquires whether the light 32 of the vehicle 10 is in the on state or in the off state, and inputs it to the arithmetic device 17. The light of the vehicle 10 is not limited to a headlight, but may be a Lo beam, a Hi beam, a blinker, a hazard lamp, or the like.

  Referring to FIG. 4 above, if an object separated by a separation distance Da at time t = 0 is detected, if the object moves at a constant speed S1 at time t = 0, the object to be measured at time t = 1 Is a separation distance Db (= Da−S1 · t). At this time, if the spectrum waveform of the measurement target is made constant regardless of the distance by using a correction map prepared in advance for each distance, the distance from the measurement target as an initial value is required, but from now on, Spectral recognition independent of the distance is also possible.

  Therefore, the correction data storage unit 18 holds a correction map 18M classified as shown in FIG. 7 based on the separation distance between the spectrum sensor 14 and the measurement target. This correction map 18M is a so-called active state in which the distance between the spectrum sensor 14 and the measurement target is a so-called passive state in which the ambient light is sunlight or the ambient light is the light 32. Whether it is a state or a measurement object. And the correction coefficient which should correct | amend the spectrum waveform of the observation light corresponding to these conditions is set for every predetermined wavelength interval in a predetermined wavelength band. That is, the correction coefficient is set every 5 (nm) between 400 and 2000 (nm). For convenience of explanation, only a part of the correction coefficients is described in the correction map 18M shown in FIG. 7, and description of other correction coefficients is omitted.

  More specifically, the correction map 18M has a wavelength of 400 (nm) as a correction coefficient when the measurement target is “person”, the ambient light is in a passive state, and the separation distance is 100 to 80 (m). 1.05 at a wavelength 405 (nm), 1.00 at a wavelength 410 (nm),..., 1.12 at a wavelength 2000 (nm). Similarly, when the separation distance is 80 to 60 (m), the correction coefficients are 0.81 at a wavelength of 400 (nm), 0.79 at a wavelength of 405 (nm), 0.60 at a wavelength of 410 (nm), ..., 0.91 at a wavelength of 2000 (nm). Further, when the separation distance is 60 to 40 (m), the correction coefficient is 0.60 at the wavelength 400 (nm), 0.59 at the wavelength 405 (nm), 0.45 at the wavelength 410 (nm),. .. 0.70 at a wavelength of 2000 (nm). Further, when the separation distance is 40 to 20 (m), the correction coefficient is 0.42 at the wavelength 400 (nm), 0.39 at the wavelength 405 (nm), 0.31 at the wavelength 410 (nm),. .. and 0.52 at a wavelength of 2000 (nm). Further, when the separation distance is 20 to 0 (m), the correction coefficient is 0.31 at the wavelength 400 (nm), 0.27 at the wavelength 405 (nm), 0.22 at the wavelength 410 (nm),. .. and 0.41 at a wavelength of 2000 (nm).

The correction coefficient when the measurement target is “human”, the observation light is in the active state, and the separation distance between the spectrum sensor 14 and the measurement target is 100 to 80 (m), the wavelength is 400 (nm). 1.83 at a wavelength 405 (nm), 1.80 at a wavelength 410 (nm),..., 2.11 at a wavelength 2000 (nm). Similarly, when the separation distance is 80 to 60 (m), the correction coefficient is 1.50 at the wavelength 400 (nm), 1.43 at the wavelength 405 (nm), 1.39 at the wavelength 410 (nm), ..., 1.80 at a wavelength of 2000 (nm).

  Next, correction of the spectrum shape using the correction map 18M will be described with reference to FIG. FIGS. 8A and 8B are explanatory diagrams for explaining an example of correction of a spectrum waveform, where FIG. 8A is a diagram illustrating a state in which ambient light is passive, and FIG. 8B is a diagram illustrating a state in which ambient light is active. c) A diagram showing an example of a spectrum waveform of a light 32, and (d) a diagram showing an example of a corrected spectrum waveform.

  In FIG. 8A, the vehicle 10 detects the spectrum waveform L40 of the observation light of the person 31 under a so-called passive condition where the environmental light is sunlight. At this time, a correction coefficient for correcting the spectrum waveform L40 of the observation light detected by the spectrum sensor 14 is acquired from the correction map 18M based on the distance to the measurement object, the passive state, and the condition of the measurement object “person”. Each light intensity is corrected by the acquired correction value, and the corrected spectrum waveform L43 of the observation light shown in FIG. 8D is calculated. Further, by comparing the spectrum waveform L43 with the spectrum shape of the dictionary data, it may be identified that the measurement object is “person”.

  In FIG. 8B, the vehicle 10 turns on the light 32, and the light is irradiated to the person 31 to be measured. Thereby, the vehicle 10 detects the spectrum waveform L41 of the observation light of the person 31 under the so-called active condition in which the environmental light is sunlight and the light 32. At this time, the spectral waveform L41 of the observation light includes a correction value based on conditions of the distance to the measurement target, the measurement target “person” and the passive state, the distance to the measurement target, the measurement target “person”, and the active state. The correction values based on the conditions are acquired from the correction map 18M, and each light intensity is corrected by the correction values, thereby obtaining the corrected spectrum shape of the observation light. Note that the spectral shape of the light 32 remains in the spectral shape of the observation light obtained at this time. Therefore, the spectrum shape of the corrected observation light is corrected by, for example, dividing by the spectrum waveform L42 set in advance as the spectrum shape of the light 32 to remove the influence of the spectrum shape of the light 32 light. Then, the spectrum waveform L43 of the observation light shown in FIG. 8D is calculated. Further, by comparing the spectrum waveform L43 with the spectrum shape of the dictionary data, it may be identified that the measurement object is “person”. Further, if the correction value in the active state is set as a correction coefficient corresponding to the spectrum waveform of the light irradiated by the light 32, the influence of the light 32 lighting generated on the spectrum shape of the observation light is determined by the correction value. It can also be corrected.

  It should be noted that, under a passive condition where the ambient light is the light of the light 32 of the own vehicle 10, the ambient light attenuates the spectrum intensity in inverse proportion to the fourth power of the distance, so the correction value under the passive condition is Compared with the correction value under the active condition, the attenuation of the light intensity is greatly compensated.

  As a result, even if a change in the spectrum waveform of the observation light obtained from the spectrum sensor 14 due to the distance to the measurement object, that is, the moving distance of the vehicle occurs, the influence of such a change is suppressed and the identification accuracy is maintained high. To be able to.

  As described above, the moving body spectrum measuring apparatus of the present embodiment can obtain the same or equivalent effects as the effects (1) to (7) of the first embodiment, and the following. Such effects can be obtained.

(8) The main factor that changes the spectrum waveform according to the distance to the object to be measured is the attenuation by the atmosphere, and the spectrum is inversely proportional to the square of the distance under the passive condition where the ambient light is not sunlight. The intensity of the spectrum is attenuated in inverse proportion to the fourth power of the distance under an active condition where the intensity is attenuated and the ambient light is not a light lit from the moving body itself. Therefore, as described above, if the correction of the spectrum waveform described above is performed so as to compensate for the attenuation of the spectrum intensity correlated with the distance between the moving body and the measurement target due to the atmosphere, the accuracy of the correction is also maintained high. Will come to be.

  (9) If the distance to the measurement object can be obtained as a relative distance to the vehicle 10, more accurate correction can be made for the attenuation of the spectrum intensity caused by the atmosphere. In this case, as in the above configuration, if the spectrum intensity is corrected in such a manner that a constant spectrum intensity is always obtained with respect to the change in the relative distance with respect to the measurement object or the vehicle 10, It becomes possible to obtain a spectrum waveform that does not depend on a change in the distance to the measurement object. That is, here, when the vehicle 10 is used as a reference, a larger correction (correction to reduce) is performed as the relative distance between the measurement object and the moving object becomes shorter, and conversely, when the measurement object side is used as a reference. As the relative distance between the measurement object and the moving object becomes longer, a larger correction (enlargement correction) is performed. Note that it is more desirable to use the mobile object as a reference if the measurement object is identified (tracked) from the spectrum waveform based on comparison with the dictionary data.

  (10) When the vehicle state acquisition device 19 is a device that can acquire the distance information from the moving body to the measurement target, it is determined according to the acquired distance information as in the above configuration. It is desirable to map the amount to be corrected as a correction coefficient in the correction data storage unit. This makes it possible to obtain a correction coefficient corresponding to the distance information in each case through a simple map calculation. Examples of the vehicle state acquisition device 19 that can acquire the distance information include a stereo camera and a radar device.

  (11) If the ambient light is not sunlight, the spectrum intensity attenuates in inverse proportion to the square of the distance under passive conditions, but the ambient light does not seem to be the light 32 lit from the vehicle 10 itself. Under active conditions, the spectral intensity attenuates in inverse proportion to the fourth power of the distance. Therefore, as described above, the spectral waveform correction factors described above are prepared separately for these passive conditions and active conditions, or for conditions that do not include an active state and conditions that include an active state. If this is done, it is possible to correct the influence of the ambient light.

  (12) Since the spectral intensity attenuates in inverse proportion to the fourth power of the distance at least under active conditions, the correction for the presence of the light lamp is the correction coefficient for the absence of the light lamp as in the above-described configuration. By mapping the correction coefficient in such a manner that the correction amount is larger than the correction, the correction accuracy can be accurately maintained.

(Third embodiment)
Next, a third embodiment of the movable body spectrum measuring apparatus according to the present invention will be described with reference to FIGS. The present embodiment is the same in configuration as the first embodiment, and the correction map as correction data held in the correction data storage unit 18 is different. In the following, the difference in the correction map will be described mainly with respect to the difference from the first embodiment.

  FIGS. 9A and 9B are diagrams illustrating a state of change in the distance of the measurement target, where FIG. 9A is a state diagram in which the distance is long, and FIG. 9B is a state diagram in the case where the distance is close. FIG. 10 is a graph showing the spectrum waveform of the observation light that changes in accordance with the change in the distance to be measured, where (a) shows the case where the change in the distance is large and (b) shows the case where the change in the distance is small. .

As shown in FIG. 9, imaging data (spectrum data) detected by the spectrum sensor 14.
This range is the range shown in the image 20. At this time, the image 20 includes an oncoming vehicle 22 that runs on the opposite lane with the road surface 21. The spectrum sensor 14 scans, for example, each line defined by dividing the region of the image 20 into a plurality of parts in the vertical direction from the bottom to the top. Spectral data is obtained for each section divided into shapes. The image 20 shows one section 23 among the sections divided in a matrix. Since the section 23 has a predetermined size regardless of the distance of the measurement target, the region of the measurement target included in the section 23 changes depending on the distance to the measurement target, and the observation light is measured in the measurement target. The spectral shape of the observation light also changes, for example, when the composition ratio of the radiation surface that emits the light changes.

  That is, when the oncoming vehicle 22 as the measurement target is far, the section 23 includes most of the oncoming vehicle 22, but conversely, when the oncoming vehicle 22 as the measurement target is close, the section 23 is opposed to the section 23. Only a part of the car 22 is included. When most of the oncoming vehicle 22 is included in the section 23 shown in FIG. 10A, the spectrum waveform L20 is detected as the observation light. At this time, when the relative speed is high and the difference in the optical path length of the observation light becomes large, for example, a spectrum waveform L21 is detected as the observation light to be acquired next. At this time, the spectrum waveform L21 is usually approximated by the spectrum shape of the dictionary data of “car” as the measurement target. That is, the spectrum waveform L20 that intersects with the spectrum waveform L21 is not only affected by attenuation due to the atmosphere, but also due to, for example, noise mixing due to a long distance or the influence of the distribution of each radiation surface in the section 23. Its shape is irregularly changed. That is, in the spectrum shape, an irregular change that is not based on the distance to the measurement target as shown by the difference ΔD21 in FIG. 10A may occur. Even in such a case, the correction map 18M or the like Thus, by setting an appropriate correction coefficient, the spectrum waveform L20 as the observation light can be appropriately corrected.

  At this time, when the relative speed is slow and the difference in the optical path length of the observation light is small, the spectral waveform L22 of the observation light as shown in FIG. 10B is detected next to the spectral waveform L20. In the same way as before, the spectrum waveforms L20 and L22 may be irregularly changed regardless of the distance, but in this case as well, if a correction coefficient for correcting the difference ΔD22 is set, both spectrum waveforms L20 will be set. , L22 can be tracked.

  In any case, if the correction coefficient is set so as to approximate the spectrum shape of the dictionary data, the measurement object can be identified and tracked by comparison with the spectrum shape of the dictionary data.

  As described above, the moving body spectrum measuring apparatus of the present embodiment can obtain an effect equivalent to or equivalent to the effects (1) to (12) of the first embodiment, and Such effects can be obtained.

  (13) The correction value of the correction map is determined so as to compensate for changes caused in the spectrum waveform due to factors other than attenuation due to the atmosphere based on prior simulations, empirical values, or experimental values. Thus, it becomes possible to correct the spectrum waveform using such a correction map.

In addition, each said embodiment can also be implemented in the following aspects, for example.
In each of the above embodiments, the case where the wavelength resolution or the wavelength region for detecting the spectrum data is not set in the spectrum sensor 14 is illustrated. However, the present invention is not limited to this, and wavelength resolution and a wavelength region may be set in the spectrum sensor. In other words, a spectrum sensor that can change the wavelength at which light intensity information is detected is adopted, and the wavelength resolution and wavelength band are set there, for example, so that the same wavelength resolution and wavelength band are specified. Only the light intensity information of the wavelength to be processed may be detected. In such a case, as shown in FIG. 11, the spectral analysis process has a spectral shape of the observation light by providing a step (step S21 in FIG. 11) for setting the wavelength in the spectral sensor. The amount of data can be reduced. The amount of data is reduced by the amount of calculation in each step of subsequent spectrum acquisition (step S22 in FIG. 11), spectrum waveform correction (step S23 in FIG. 11), and spectrum analysis (tracking process) (step S24 in FIG. 11). It is also possible to reduce the processing time required for the spectrum analysis process by reducing. As a result, even when such a spectrum measuring apparatus is mounted on a vehicle as a moving body, the processing capability in real time is improved, and the applicability to a vehicle having a high moving speed is also increased. .

  In each of the above embodiments, the case where the correction data of the spectrum measuring apparatus 11 is held in the form of a map has been illustrated. However, the present invention is not limited to this, and the correction data may be held in the form of a mathematical expression or the like. In this case, the correction value may be obtained by each calculation. As a mathematical expression indicating the correction value, for example, adoption of a quadratic function is conceivable under a passive condition, and adoption of a quaternary function is conceivable under an active condition.

  In each of the above embodiments, the case where correction data is held in the correction data storage unit 18 is illustrated. However, the present invention is not limited to this, and the correction data may be held in other storage areas, such as a dictionary data storage unit or an internal memory of the arithmetic device. If the configuration without the correction data storage unit 18 is adopted, the configuration of the spectrum measuring apparatus 11 can be facilitated.

  -The vehicle of each said embodiment may be a motor vehicle. With such a spectrum measurement device, even when mounted on a car, the recognition and tracking processing of the measurement object that sequentially approaches as the vehicle runs on the road is performed in real time so that appropriate driving support can be performed. Become. As a result, the possibility of adopting the spectrum measuring apparatus for an automobile is increased.

  In addition, each said embodiment is not restricted to the motor vehicle as a vehicle, As long as it is a mobile body which moves on a road, such as a motorcycle and a robot, adoption of such a spectrum measuring apparatus for mobile bodies is possible.

  DESCRIPTION OF SYMBOLS 10 ... Vehicle, 11 ... Spectrum measuring device, 12 ... Human machine interface, 13 ... Vehicle control device, 14 ... Spectrum sensor, 15 ... Spectrum data processing device, 16 ... Dictionary data storage part, 17 ... Calculation device, 18 ... Correction data Storage unit, 18M ... correction map, 19 ... vehicle state acquisition device as moving state acquisition device, 20 ... image, 21 ... road surface, 22 ... oncoming vehicle, 23 ... section, 30 ... sun, 31 ... person, 32 ... light.

Claims (9)

Spectral measurement for mobile objects equipped with a spectrum sensor capable of measuring wavelength information and light intensity information and identifying measurement objects around the mobile object based on the spectrum waveform of the observation light detected by this spectrum sensor A device,
A dictionary data storage unit in which a spectrum waveform including wavelength information and light intensity information for a plurality of predetermined measurement objects is stored as dictionary data;
An arithmetic unit for identifying the measurement object based on a comparison operation between the spectral waveform of the observation light and the spectral waveform stored in the dictionary data storage unit;
A movement state acquisition device that acquires information regarding the movement state of the mobile body,
The arithmetic device is configured to attenuate the spectrum intensity of the observation light, which correlates the spectrum waveform of the observation light with the distance to the measurement object, based on information about the movement state of the moving body acquired by the movement state acquisition device. To compensate for the spectral waveform stored in the dictionary data storage unit with correction to compensate ,
A mobile spectrum measuring apparatus, wherein a correction value used for the correction is made different depending on the presence or absence of a light lamp . The distance to the measurement object is a relative distance to the moving object, and the correction is performed in such a manner that a constant spectral intensity is always obtained with respect to the change of the relative distance with respect to the measurement object or the moving object. The mobile spectrum measuring apparatus according to claim 1 . The information regarding the moving state of the moving body acquired by the moving state acquisition device is distance information of the moving body with respect to the measurement target, and the amount to be corrected is determined according to the acquired distance information for each time. The mobile body spectrum measuring apparatus according to claim 2 , which is mapped in a correction data storage unit as a correction coefficient. 4. The mobile spectrum measuring apparatus according to claim 3 , wherein the correction coefficient is mapped with a plurality of values for each distance information for each wavelength band having a different spectral waveform. Information relating to the presence or absence of light lighting of the moving body is also acquired as information relating to the movement state of the moving body acquired by the movement state acquisition device, and the correction coefficient is a different correction according to the presence or absence of the light lighting. 5. The moving body spectrum measuring apparatus according to claim 3, wherein the moving data spectrum is mapped to the correction data storage unit as a coefficient. The mobile spectrum measuring apparatus according to claim 5 , wherein the correction coefficient is mapped in such a manner that a correction for the presence of light lighting is a correction amount larger than a correction for the absence of light lighting. 5. The mobile spectrum measuring apparatus according to claim 3 , wherein the correction coefficient is mapped to the correction data storage unit as a different correction coefficient corresponding to a spectrum waveform of a light irradiated by a light. Spectral measurement for mobile objects equipped with a spectrum sensor capable of measuring wavelength information and light intensity information and identifying measurement objects around the mobile object based on the spectrum waveform of the observation light detected by this spectrum sensor A device,
A dictionary data storage unit in which a spectrum waveform including wavelength information and light intensity information for a plurality of predetermined measurement objects is stored as dictionary data;
An arithmetic unit for identifying the measurement object based on a comparison operation between the spectral waveform of the observation light and the spectral waveform stored in the dictionary data storage unit;
A movement state acquisition device that acquires information regarding the movement state of the mobile body,
The arithmetic device is configured to attenuate the spectrum intensity of the observation light, which correlates the spectrum waveform of the observation light with the distance to the measurement object, based on information about the movement state of the moving body acquired by the movement state acquisition device. and a correction to perform a comparison operation between the spectrum waveform stored in the dictionary data storage unit processing to compensate, etc. distance the mobile body you move a short Ho within a predetermined time small fence, the distance the correction amount is largely long nearly as, after correcting one of the spectral waveform of the observation light to the front and rear by the predetermined time, based on a comparison of the corrected ones and the other, the observation light is obtained by the front and rear by the predetermined time And a process for determining the identity of the measured object . The spectrum measuring apparatus for a moving body according to claim 8, wherein a correction value used for the correction is made different according to the presence or absence of a light lamp .

Images