JPH06195464A - Detector of object - Google Patents

Abstract

PURPOSE:To make it possible to surely detect an object even when it overlaps partially by making the feature information on the object into a locus and detecting the object from the locus. CONSTITUTION:This detector is provided with a monitoring camera 1 performing an image pickup for an object, an image memory 5 where the video signals of the object outputted from this monitoring camera 1 are successively supplied by a screen unit and the video signals are successively recorded by overlapping them and a feature extraction part 4 extracting the feature information of the object from the video signal recorded in this image memory 5 and making the video signals into a locus. The partial feature information is extracted from the object imaged by the monitoring camera 1 and it is made into a locus, whether the locus extends from the end edge of the monitoring area of the monitoring camera 1 or not is detected, and when the locus extends from the end edge, it is detected as the object which enters the monitoring area newly. Therefore, the object can be surely detected even locus plural objects overlap partially because the object is detected by the partial feature information.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a target object detecting apparatus for detecting a target object using an image picked up by a surveillance camera.

[0002]

2. Description of the Related Art Conventionally, when a target object is detected by using an image picked up by a surveillance camera, the picked-up image signal is taken in, for example, one frame unit and the difference between the target object and the background is taken. The target object is detected by the method.

[0003]

The above-mentioned conventional target object detection apparatus detects one target object as a whole. Therefore, even if there are a plurality of target objects, it may be determined as one target object if the plurality of target objects partially overlap.

Further, when the difference between the target object and the background is taken in order to detect the target object, it may be difficult to separate the noise and the target object, and the target object may not be correctly identified.

According to the present invention, a part of a target object is converted into a trajectory as feature information, and the trajectory is used to detect the target object. It is therefore an object of the present invention to provide a target object detection apparatus capable of reliably detecting the target object. To aim.

[0006]

A target object detecting apparatus according to the present invention comprises a surveillance camera for picking up an image of a target object traveling in a predetermined surveillance area, and a video signal output from the surveillance camera in order on a screen-by-screen basis. The feature extraction unit that is supplied and extracts the feature information of the target object for each screen unit, and the feature information of the target object that is extracted by this feature extraction unit are sequentially recorded over a predetermined time longer than one screen unit. An image memory for converting the characteristic information into a locus, and means for detecting whether or not the locus of the characteristic information converted into a locus in the image memory extends from the edge of the surveillance area of the surveillance camera. .

Further, the target object detecting apparatus of the present invention supplies a monitoring camera for picking up an image of a target object traveling in a predetermined monitoring area, and a video signal of the target object output from the monitoring camera sequentially in units of screens. An image memory for sequentially recording video signals of the target object over a predetermined time longer than one screen unit, and characteristic information of the target object based on the video signals recorded in the image memory for the predetermined time. And a means for detecting whether or not the trajectory of the characteristic information traced by the image memory extends from the edge of the surveillance area of the surveillance camera. It has.

[0008]

In the present invention, partial characteristic information is extracted from the target object imaged by the surveillance camera, and the characteristic information is made into a locus. Then, it is detected whether or not the trajectory of the characteristic information extends from the edge of the surveillance area of the surveillance camera, and if it extends from the edge of the surveillance area, it is detected as a target object newly entering the surveillance area.

Therefore, since the target object is detected by the partial characteristic information, the target object can be detected even if a plurality of target objects partially overlap each other.

Further, when the trajectory of the characteristic information extends from the edge of the monitoring area, it is determined that the target object has newly entered the monitoring area. Therefore, the target object can be detected with certainty.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below with reference to FIG.

As shown in FIG. 2, the monitoring camera 1 is arranged downward so that a target object running on a road 21, that is, a vehicle 22 can be imaged.
In FIG. 2, two vehicles 22 are shown on the road 21.

In the surveillance camera 1, for example, as shown in FIG. 3, the shoulder 31 of the right lane, the center line 32, the shoulder 33 of the left lane, etc. are imaged in the surveillance area.

The image taken by the surveillance camera 1 (FIG. 1) is output from the surveillance camera 1 as a video signal. CP
Under the control of U2, the video signal is taken in by the AD converter 3 and converted into a digital signal. Then, it is supplied to the feature extraction unit 4. The video signal whose features are extracted by the feature extraction unit 4 is recorded in the image memory 5. The image memory 5
Is composed of, for example, first and second two memories 5a and 5b.

A method of processing the video signal input to the feature extraction unit 4 and a method of recording the processed video signal in the image memory 5 will be described with reference to the flowchart of FIG.

The video signal converted into a digital signal by the A / D converter 3 is controlled by the CPU 2 in step 41.
Then, one screen unit, for example, one field of the video signal is taken into the feature extraction unit 4. First, the video signal of the first field is captured.

The video signal fetched by the feature extraction unit 4 is
Under the control of the CPU 2, a part of the target object is extracted as the characteristic information in step 42.

For example, the characteristic information F1 is extracted from the first field video signal.

As the characteristic information, when the target object is, for example, a vehicle during night driving, a portion having a higher brightness than other portions such as a tail lamp and a head lamp is selected. Therefore, the characteristic information of the target object is extracted using the difference in brightness.

There are a plurality of high-luminance portions such as tail lamps and headlamps in one target object.
Therefore, usually, a plurality of pieces of characteristic information are extracted from one target object.

However, in the following embodiments, for convenience of explanation,
It is assumed that one feature information is extracted from one target object.

The characteristic information F1 includes the positional information in the surveillance area of the surveillance camera 1 where the characteristic information F1 is picked up,
In step 43, it is recorded in the first memory 5a.

Next, at step 44, the video signals of the second and subsequent fields are sequentially fetched.

For example, the video signal of the second field is taken in, and in step 45 the characteristic information F2 corresponding to the tail lamp or head lamp is extracted. This feature information F
2 is recorded in the second memory 5b in step 46.

Characteristic information F recorded in the second memory 5b
2 is subjected to arithmetic processing in step 47 and added to the first memory 5a. Therefore, at this time, the characteristic information F1 and the characteristic information F2 are recorded in the first memory 5a.

Then, in step 48, the signal processing from step 44 to step 47 is repeated for a fixed time T for the video signals of the third and subsequent fields.

It is assumed that K field image signals arrive during the fixed time T and, for example, one piece of characteristic information is detected from all field image signals. In this case, when the fixed time T has elapsed, K pieces of characteristic information F1 to FK are recorded in the first memory 5a.

Characteristic information F recorded in the first memory 5a
Each of 1 to FK has a time interval corresponding to one field. Therefore, K pieces of characteristic information F1 to FK
If the tail lamps are connected to each other, the trail of the tail lamp and the like will be shown for a certain time T as the vehicle moves.

Then, after a lapse of a fixed time T, the characteristic information F1 to FK recorded in the first memory 5a is output.

Therefore, the characteristic information is output from the first memory 5a at regular time intervals T. The characteristic information output from the first memory 5a at regular time intervals T is transferred to the trajectory storage unit 7 through the trajectory processing calculation unit 6 in FIG. 1 and recorded there. At that time, the final characteristic information, for example, the position information of the Kth characteristic information FK is also recorded at each constant time T.

When the characteristic information recorded in the first memory 5a is transferred to the locus storage unit 7, the image memory 5, such as the first and second memories 5a and 5b, is stored before that.
The characteristic information and the like recorded in (4) are deleted.

Thereafter, the same signal processing as described above is repeated for the fixed time T for the next field video signal, for example, the K + 1th field video signal and thereafter.

When the fixed time starts, the first field
For the feature information FK + 1 of the field image signal, for example, the K + 1th field image signal, the position information of the surveillance area of the surveillance camera is simultaneously detected.

As described above, the first field where the fixed time starts
The position information of the feature information FK + 1 detected in the field image signal, for example, the K + 1th field image signal is obtained. Then, it is collated with the position information of the characteristic information FK detected in the last, for example, the Kth field image signal of the last fixed time.

As a result, for example, when the position information of both is close to each other, the characteristic information FK and FK + 1 of both are judged to be the loci of the same vehicle. The determination as to whether or not the loci of the same vehicle are made is performed every time a certain period of time elapses.

In addition, K pieces of image data output from the image memory 5, for example, the first memory 5a, at constant time intervals T (when characteristic information is detected from each field, there are fields where characteristic information is not detected). If so, the number of feature information is reduced accordingly.) The feature information is added to the DA converter 8 and converted into an analog signal.

The characteristic information converted into the analog signal is reproduced as an image on the display unit 9. On the screen of the display unit 9, as shown in FIG. 5, a locus 51 is displayed as, for example, two tail lamp loci 51a and 51b.

The image reproduced on the display unit 9 is not limited to the characteristic information output from the image memory 5. For example, the characteristic information recorded in the trajectory storage unit 7 can be converted into an analog signal by the DA converter 8 and reproduced.

The trajectory of the characteristic information of the vehicle, which has been converted into a trajectory as described above, is tracked by the trajectory processing operation unit 6 (FIG. 1).

An example of trajectory tracking performed by the trajectory processing calculation unit 6 will be described below with reference to the flowchart of FIG.

For example, in step 61, a new vehicle is detected.

Assuming that the outer frame of the surveillance area of the surveillance camera 1 corresponds to, for example, the quadrangular detection line 52 of FIG. 5, the locus 51 of the characteristic information is the detection line 52 as shown in FIG.
If it intersects with the new vehicle, it is determined that the new vehicle has entered the surveillance area of the surveillance camera 1. In this way, a new vehicle is detected.

If the locus of the characteristic information does not intersect the detection line, it is excluded as unnecessary information such as noise.

When the vehicle is approached from any direction, the detection line is provided so as to draw a closed loop along the outer edge of the monitoring area.

Further, in step 62, the characteristic information FK of the first field video signal for which a fixed time starts, in the case of the above embodiment, for example, the K + 1th field video signal.
Location information is recalled for +1.

In step 63, the last field video signal, for example, the Kth field video in the case of the above embodiment, which was called in step 62 but is a predetermined time T immediately before that, is called. The position information of the signal characteristic information FK is called. In step 64, the signal processing of steps 62 and 63 is performed on the characteristic information of all the vehicles.

Then, the position information corresponding to the characteristic information of the first field video signal, which is called in step 62 and starts for the fixed time, and the last field of the previous fixed time T, which is called in step 53. The collation with the position information for the characteristic information of the video signal is performed for all trajectories. As a result, when the position information of both is close, both are processed as the continuous same vehicle loci.

Then, in step 65, when the loci of the same continuous vehicle are displayed, the loci are displayed as a continuous locus.

By the processing of the locus as described above, the traveling state of the vehicle is observed, for example, as follows.

As shown in FIG. 7, when many vehicle loci 71 cross the center line 72 at a certain point from the left lane to the right lane, or in the opposite direction from the right lane to the left lane, the roads before and after the center line 72. It is suggested that an abnormal vehicle such as a low speed vehicle or a stopped vehicle may exist in the shaded area 73. In FIG. 7, the locus of one vehicle is shown as a representative. Moreover, 74 is a road shoulder.

When the vehicle characteristic information is obtained from the tail lamps, two loci 81a and 81b are obtained from one vehicle as shown in FIG. In this case, even if a plurality of vehicles overlap due to traffic jam or the like, and the locus 81b on one side disappears on the way, the locus can be traced using the other locus 81a.

In the first embodiment, the video signal output from the surveillance camera is converted into a digital signal,
After that, the features are extracted, and the extracted feature information is recorded in the image memory.

However, the same effect can be obtained by a configuration in which the video signal output from the surveillance camera is recorded in the image memory for a certain period of time and then the feature is extracted.

Here, as a second embodiment of the present invention, a case of recording a video signal in an image memory and then extracting features will be described with reference to FIG. In FIG. 9, the same parts as those in FIG. 1 showing the first embodiment are designated by the same reference numerals.

The image picked up by the surveillance camera 1 is output from the surveillance camera 1 as a video signal. Under the control of the CPU 2, the video signal is taken into the AD converter 3 and converted into a digital signal. The video signal converted into the digital signal is sequentially recorded in the image memory 5 for one screen unit, for example, for each field of the video signal, over a fixed time. The image memory 5 is composed of, for example, first and second memories 5a and 5b.

Here, a method of recording the video signal converted into the digital signal in the image memory 5 will be described with reference to the flowchart of FIG.

Under the control of the CPU 2, step 101
Then, the video signal of the first field is taken into the AD converter 3.

The video signal taken into the AD converter 3 is converted into a digital signal and recorded in the first memory 5a in step 102.

Next, at step 103, the video signals of the second and subsequent fields are sequentially fetched.

For example, the video signal of the second field is A
-It is taken in by the D converter 3 and converted into a digital signal,
In step 104, it is recorded in the second memory 5b.

The video signal recorded in the second memory 5b is arithmetically processed in step 105, and the first memory 5a is processed.
In addition, it is recorded by being superposed on the video signal already stored in step 102. Therefore, at this time, the video signals of two fields are recorded in the first memory 5a so as to overlap each other.

Then, in step 106, for the video signals of the third and subsequent fields, the steps 103,
The signal processing of 104 and 105 is repeated for a fixed time T.

When k field image signals arrive during the fixed time T, at the time when the fixed time T has passed, the first
In the memory 5a, video signals for k fields are recorded in an overlapping manner.

Then, after the elapse of the predetermined time T, the video signals for the k fields which are overlapped and recorded in the first memory 5a are output.

This video signal is supplied to the characteristic extraction unit 4 (FIG. 9) under the control of the CPU 2, and the characteristic information is extracted in step 107.

The feature extraction unit 4 extracts the feature information from the video signals in which k pieces are overlapped by the same method as in the first embodiment. As a result, k fields, that is, the constant time T
The feature information between the two is extracted. At this time, CPU2
Of the first and last field video signals at each constant time T, for example, the first and k
Position information is added to the th characteristic information f1 and fk.

The characteristic information output from the characteristic extracting unit 4 at constant time intervals T is supplied to the locus processing calculating unit 6 in FIG.
The same trajectory processing as that of the first embodiment is performed. After that, the characteristic information is transferred from the trajectory processing calculation unit 6 to the trajectory storage unit 7 and recorded there.

When the characteristic information is transferred to the locus storage unit 7, the video signals previously recorded in the image memory 5, for example, the first and second memories 5a and 5b are erased. Thereafter, the same signal processing as described above is repeated for the fixed time T for the next field video signal, for example, the k + 1th field video signal and thereafter.

The position information of the characteristic information fk + 1 of the first field video signal, for example, the k + 1th field video signal at the beginning of the fixed time, and the last, for example, the kth field of the fixed time before that. Feature information fk of video signal
The configuration is the same as that of the first embodiment, in which the characteristic information fk and fk + 1 of both are determined to be the locus of the same vehicle when the position information of both is close to each other. The determination as to whether or not the loci of the same vehicle are made is performed every time a certain period of time elapses.

The video signal recorded in the image memory 5 can be converted into an analog signal by the DA converter 8 and reproduced as an image on the display unit 9.

The structure for displaying the extracted feature information on the display unit and the feature information recorded in the trajectory storage unit are
Similar to the first embodiment, the configuration in which the -A converter converts into an analog signal and displays it on the display unit, and the configuration in which the trajectory processing calculation unit tracks the trajectory from the characteristic information of the vehicle that has been transformed into the trajectory is the same as the first embodiment. Therefore, the description is omitted.

In the case of the second embodiment, since the number of feature extraction processes is reduced, the processing speed can be increased.

In each of the above-described embodiments, the image memory is composed of two memories, but the image memory may be composed of one memory.

Further, although the screen unit for extracting the characteristic information is one field, it can also be performed for each frame.

Further, in each of the above-described embodiments, the trajectory of the characteristic information is segmented at regular intervals. By dividing the time for forming the trajectory in this way, the moving speed of the target object such as a vehicle can be known from the length of the trajectory formed in a certain time.

Further, since the target object is detected not as the whole target object but as a part of the target object as feature information by utilizing the difference in brightness, the target object can be detected reliably, and a plurality of target objects partially overlap each other. However, the target object can be detected.

[0077]

According to the present invention, the characteristic information of the target object is converted into a trajectory, and the target object is detected from the trajectory. Therefore, the target object can be reliably detected even if the target objects partially overlap.

[Brief description of drawings]

FIG. 1 is a circuit configuration diagram illustrating a first embodiment of the present invention.

FIG. 2 is a view for explaining the present invention and is a schematic diagram showing the arrangement of surveillance cameras.

FIG. 3 is a diagram for explaining the present invention, and is a diagram showing a pattern captured by a surveillance camera.

FIG. 4 is a flowchart illustrating the operation of the present invention.

FIG. 5 is a diagram for explaining the operation of the present invention, and is a diagram showing a locus displayed on a display.

FIG. 6 is a flowchart showing the operation of the present invention.

FIG. 7 is a diagram for explaining the operation of the present invention and is a diagram showing a road condition displayed by a display device.

FIG. 8 is a diagram for explaining the operation of the present invention, and is a diagram showing a locus displayed on a display.

FIG. 9 is a circuit configuration diagram illustrating a second embodiment of the present invention.

FIG. 10 is a flowchart for explaining the operation of the second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Surveillance camera 2 ... CPU 3 ... A-D converter 4 ... Feature extraction part 5 ... Image memory 6 ... Locus processing calculation part 7 ... Locus storage part 8 ... DA converter 9 ... Display part

Claims (2)

[Claims] 1. A surveillance camera for picking up an image of a target object traveling in a predetermined surveillance area, and a video signal of the target object output from the surveillance camera are sequentially supplied in units of screens. A feature extracting means for extracting the feature information, an image memory for sequentially recording the feature information of the target object extracted by the feature extracting means for a predetermined time longer than one screen unit, and converting the feature information into a trajectory, A device for detecting a target object, comprising: means for detecting whether or not the trajectory of the characteristic information traced by the image memory extends from the edge of the surveillance area of the surveillance camera. 2. A surveillance camera for picking up an image of a target object traveling in a predetermined surveillance area, and video signals of the target object output from the surveillance camera are sequentially supplied on a screen-by-screen basis for a predetermined time longer than one screen unit. An image memory for sequentially recording the video signals of the target object in order, a feature extraction unit for extracting the feature information of the target object from the video signals recorded in the image memory, and converting the feature information into a locus, A detection device for a target object, comprising: a means for detecting whether or not the trajectory of the feature information that has been transformed into a trajectory by the feature extraction unit extends from the edge of the surveillance area of the surveillance camera.