JP3400643B2 - In-vehicle image processing device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle-mounted image processing apparatus for recognizing a traveling environment using an image signal from a vehicle-mounted image pickup means, and more particularly to a vehicle-mounted image processing apparatus having a function of detecting an abnormality in an image processing circuit. The present invention relates to an image processing device.

[0002]

2. Description of the Related Art Conventionally, a vehicle-mounted image processing apparatus for performing image processing (edge processing and binarization processing) on image data input by a CCD camera or the like to recognize a running environment such as a white line on a road or a preceding vehicle. Have been proposed.

In this type of device, when the normal control mode is selected in response to a driver's request, a process of detecting an object around the host vehicle during steady operation (for example, a driving state at a predetermined vehicle speed or higher) is executed. Then, for example, in response to the image recognition result, constant speed traveling control or the like that automatically follows while maintaining the inter-vehicle distance from the preceding vehicle is performed.

For example, in the apparatus disclosed in Japanese Patent Laid-Open No. 3-273498, an edge image (contour image) is extracted based on the vehicle peripheral image input from the image input unit, and this edge image is binarized. By doing this, the white line position is recognized.

Further, in the device disclosed in Japanese Patent Publication No. 6-24035, an edge image is extracted based on the vehicle peripheral image input from the image input unit, and a threshold value by a look-up table is applied to the edge image. The shape of the traveling road is recognized by performing processing and Hough conversion.

A conventional on-vehicle image processing apparatus will be described below with reference to the drawings. FIG. 11 is a block diagram showing a schematic configuration of a conventional vehicle-mounted image processing apparatus disclosed in, for example, Japanese Patent Laid-Open No. 7-244717. In the figure, 1 is a CCD camera as an image pickup means and 2 is a CCD camera 1. The electronic control unit, or ECU, includes a microcomputer for processing a captured image signal A.

The ECU 2 includes a CPU 3 having various image processing arithmetic functions and an image signal A from the CCD camera 1.
An amplifier 10 for amplifying the image signal, an A / D converter 11 for converting the amplified image signal A into a digital signal, an image memory 12 for storing the digitally converted image signal A as image data D, and an image data D. CP after image processing
The image processing circuit 13 for inputting to U3.

The image processing circuit 13 performs various kinds of image processing (edge processing and binarization processing) on the image data D stored in the image memory 12, and inputs the processing result (binarization data) to the CPU 3. .

The CPU 3 includes a white line position detecting means 14 and a preceding vehicle position detecting means 15 for detecting the white line position on the traveling road and the position of the preceding vehicle based on the binarized data Dc via the image processing circuit 13. .

Although not shown here, the CPU 3
Includes a control mode selecting means selected by a driver's operation, and enables the processing function of each detecting means 14 and 15 when the normal control mode is selected. Further, the processing results of the detection means 14 and 15 are used, for example, in constant speed control or the like in which the subject vehicle automatically follows the preceding vehicle.

FIG. 12 is a flowchart showing the processing operation in the CPU 3, and in FIG. 12, each step S14
And S15 correspond to the processing operations of the white line position detecting means 14 and the preceding vehicle position detecting means 15 in FIG. 11, respectively. The CPU 3 detects the white line position and the preceding vehicle position on the road by further performing the white line detection processing step S14 and the preceding vehicle position detection processing step S15 on the binarized data Dc input from the image processing circuit 13. To do.

FIG. 13 is a flow chart specifically showing the edge processing and binarization processing executed by the image processing circuit 13 and the white line detection processing step S14 executed by the white line position detecting means 14.

FIG. 14 is an explanatory view showing an example of an image picked up by the CCD camera 1, 40 is an image pickup frame, and 41 is an image pickup frame.
Is one scanning line on the imaging frame 40, 42 is a character on the traveling lane, 43 is a white line on the traveling lane, and 44 is a preceding vehicle.

FIG. 15 shows a scanning line 41 in the image pickup frame 40.
FIG. 6 is an explanatory diagram showing a result of performing data processing on the upper image signal A, where the horizontal axis represents the scanning position in the imaging frame 40,
The vertical axis represents the brightness value (signal level) at each scanning position.

In FIG. 15, (a) shows the image data D corresponding to the image signal A on the scanning line 41 in FIG. 14, and (b) to (d) show the edge with respect to the image data D. The results (each data) of the processing, binarization processing, and white line detection processing (see FIG. 13) are shown.

Next, referring to FIGS. 12 to 15, FIG.
The white line detection processing operation by the conventional vehicle-mounted image processing apparatus shown in FIG. 1 will be described. First, in FIG. 13, the image processing circuit 13 extracts an edge of the captured image data D (step S20).

At this time, the edge is extracted by the image data D
It is obtained by the absolute value of the difference between the brightness levels of pixels that are adjacent to each other above and below or to the left and right. In this way, by performing the edge extraction processing (step S20) on the image data D, the image data D becomes the edge image data Db as shown in FIGS. 15A to 15B.

Then, the image processing circuit 13 extracts only the edges of a predetermined level or more from the edge image data Db based on the result of comparison with the threshold value Dbr (see the alternate long and short dash line) of the specific level. Db is shown in FIG.
The binarized data Dc as shown in (c) is obtained (step S2
1).

In this way, in the image processing circuit 13,
The binarized data Dc is obtained by performing the binarization process (step S21) on the edge image data Db obtained by the edge extraction process (step S20), and the binarized data Dc is stored in the CPU 3. Is entered.

Next, the CPU 3 performs a time integration process (step S22) on the binarized data Dc, as shown in FIG.
The integrated data Dd as shown in (d) is obtained. At this time, paying attention to one pixel in the imaging frame 40, the integrated data Dd (n) at the current time (t = n) shows the brightness value P (n) at the current time and the previous time (t = n−1). Using the integral data Dd (n-1) in and the predetermined time constant τ, it is obtained as in the following equation (1).

[0021]   Dd (n) = Dd (n−1) × (1-1 / τ) + P (n) / τ ... (1)

Here, if the time constant τ is set large, the integration will take a long time, and if the time constant τ is set small, the integration will take a short time. As described above, when the binary integration data Dc shown in FIG. 15C is subjected to the time integration processing by the equation (1), the integration data Dd as shown in FIG. 15D is obtained.

The integrated data Dd is a smoothed small value for a contour line which appears temporally unstable like the character 42 in the image pickup frame 40, and is stable in time like a white line 43. It is a large value for contour lines appearing at the same position.

Next, in order to extract the integrated data Dd of a predetermined level or higher as a white line, the integrated data Dd is compared with a threshold Ddr of a predetermined level, and the position L indicating the threshold Ddr or higher is detected.
Is detected as a white line position (step S23).

Finally, the scanning position is updated above the current scanning position, and the scanning position is updated by the imaging frame 4
It is determined whether or not the entire 0 has been completed (step S24). If it is determined that the scanning position has not been updated (that is, NO), the process returns to step S20 and the above process is repeated, and the scanning position reaches the uppermost position and the updating is finished (that is, YES). If it is determined, the processing of FIG. 13 ends.

As described above, the white line detection processing (steps S20 to S24) of FIG. 13 is repeated from the bottom to the top of the image in the image pickup frame 40 shown in FIG.
The white line position L on the scanning line can be detected.

Next, the operation of the preceding vehicle position detecting means 15 in FIG. 11 will be described with reference to the flowchart of FIG. 16 and the explanatory views of FIGS. 17 and 18. Figure 1
6 is a flow chart specifically showing the edge processing and binarization processing executed by the image processing circuit 13 and the white line detection processing step S15 (see FIG. 12) executed by the preceding vehicle position detecting means 15, S20 to S22 and S24 is the same step as described above (see FIG. 13).

FIG. 17 is an explanatory view showing an example of an image picked up by the CCD camera 1, which is 40, 41, 43 and 4.
4 is the same as that described above (see FIG. 14). FIG. 18 is an explanatory diagram showing the result of data processing of the image signal A on the scanning line 41 in the imaging frame 40, and corresponds to FIG. 15 described above.

In FIG. 18, (a) to (c) are the same as those described above (FIG. 15), and (d) shows the result of subjecting the image data D to the preceding vehicle detection process (see FIG. 16). .
First, in FIG. 16, the image processing circuit 13 performs edge extraction processing (step S20) and binarization processing (step S21) to generate edge image data Db and binarized data Dc, and the CPU 3 performs integration processing. (Step S
22) is performed to obtain the integrated data Dd.

Subsequently, the CPU 3 compares the integrated data Dd with a threshold value Ddr 'of a predetermined level in order to extract the integrated data Dd of a predetermined level or higher as a preceding vehicle candidate, and positions L1 to L3 indicating the threshold value Ddr' or higher. Is searched as a preceding vehicle candidate (step S33).

At this time, since the automobile has a predetermined width or more, the predetermined width is set in advance, and the preceding vehicle 44 is set in the area satisfying the predetermined width or more from the preceding vehicle candidates L1 to L3. Is present and is detected as the position of the preceding vehicle (step S34).

In this case, comparing the area widths of the preceding vehicle candidates L1 to L3 with the predetermined width, the preceding vehicle candidates L1 and L3 are compared.
Region width does not satisfy the condition of being equal to or larger than a predetermined width, and the preceding vehicle candidate L
Since only the region width of 2 satisfies the condition of being equal to or larger than the predetermined width, the position of the preceding vehicle candidate L2 can be determined as the preceding vehicle candidate position.

Subsequently, in step S24, it is determined whether or not the scanning position is updated above the current scanning position and the scanning is completed over the entire imaging frame 40, and the processes of steps S20 to S34 are executed until it is completed. Execute repeatedly. Then, after the scanning is completed, the position L2 of the preceding vehicle 44 is output (step S35), and the processing of FIG. 16 is completed.
If the preceding vehicle position is not detected at all the scanning positions, it is determined that the preceding vehicle 44 does not exist.

However, when an abnormality occurs in the image processing circuit 13 in the ECU 2, the binarized data Dc input to the CPU 3 is in an error state. There is a possibility that the result of the preceding vehicle detection process performed by the vehicle position detection unit 15 may be in an error state and the white line 43 and the preceding vehicle 44 may not be accurately detected.

[0035]

As described above, since the conventional vehicle-mounted image processing apparatus does not have means for detecting the abnormality occurrence of the image processing circuit 13 in the ECU 2, the abnormality occurrence of the image processing circuit 13 occurs. There is a problem that it is impossible to prevent the erroneous detection of the position information of the white line 43 and the preceding vehicle 44 because the image processing based on the normal control mode is executed even at times.

The present invention has been made to solve the above-mentioned problems, and is capable of detecting an abnormal occurrence of an image processing circuit and preventing erroneous information output when the abnormal occurrence occurs. Aim to get.

[0037]

According to a first aspect of the present invention, there is provided an in-vehicle image processing device, which is mounted on a vehicle and which produces an image signal for capturing an image of the periphery of the vehicle, and image data based on the image signal. An image processing circuit that extracts edge image data from the image and performs binarization processing, a position detection unit that detects a position of an object around the vehicle based on the binarized data from the image processing circuit, and a predetermined test pattern are stored. Based on the binarized data of the test pattern through the image processing circuit, and the switching means for switching the input signal to the image processing circuit from the image data to the test pattern in response to the control pattern state of the vehicle. Circuit abnormality detecting means for detecting abnormality in the image processing circuit by means of the circuit abnormality detecting means. Comparing a predetermined number of pixels and used as a reference value, when both match condition is not satisfied, it is to determine the abnormality of the image processing circuit.

According to a second aspect of the present invention, there is provided the vehicle-mounted image processing apparatus according to the first aspect, wherein the circuit abnormality detecting means sets the projection value, which is the sum of the respective pixel values of the edge image data based on the test pattern, to the pixel value. The projection value is obtained as a number, and the projection value is compared with a determination value that is a predetermined number of pixels, and when the projection value deviates from the allowable range based on the determination value, the abnormality of the image processing circuit is determined.

According to a third aspect of the present invention, there is provided the vehicle-mounted image processing device according to the first or second aspect, wherein the circuit abnormality detecting means is the position detecting means when the abnormality of the image processing circuit is determined. The processing function is stopped.

According to a fourth aspect of the present invention, there is provided the vehicle-mounted image processing apparatus according to any one of the first to third aspects, wherein the circuit abnormality detecting means abnormally diagnoses the timing according to the control state of the vehicle. Including the timing search means to search as a mode, when the abnormality diagnosis mode is searched,
A switching signal for operating the switching means is output.

According to a fifth aspect of the present invention, there is provided the vehicle-mounted image processing device according to any one of the first to fourth aspects, in which the peripheral object detected by the position detecting means is on the traveling lane of the vehicle. At least one of the white line and the preceding vehicle.

According to a sixth aspect of the present invention, the vehicle-mounted image processing apparatus according to the fifth aspect includes control mode selection means for selectively enabling the processing function of the position detection means to set the normal control mode. The timing search means determines the abnormality diagnosis mode when the normal control mode of the vehicle is not selected, and validates the processing function of the circuit abnormality detection means.

According to a seventh aspect of the present invention, in the on-vehicle image processing apparatus according to the sixth aspect, the timing search means controls the position detecting means so that the processing function of the position detecting means becomes invalid when the normal control mode is selected. In this state, the abnormality diagnosis mode is determined and the processing function of the circuit abnormality detection means is enabled.

According to the eighth aspect of the present invention, in the vehicle-mounted image processing apparatus according to the seventh aspect, the control state in which the processing function of the position detecting means is invalidated is when the vehicle starts.

According to a ninth aspect of the present invention, in the vehicle-mounted image processing apparatus according to the seventh aspect, the control state in which the processing function of the position detecting means is disabled is defined as a state in which the traveling speed of the vehicle is equal to or lower than a predetermined vehicle speed. It was done.

According to a tenth aspect of the present invention, there is provided a vehicle-mounted image processing apparatus according to the seventh aspect, in which the control state in which the processing function of the position detecting means is disabled is a state in which the steering angle of the vehicle is a predetermined angle or more. It was done.

[0047]

DETAILED DESCRIPTION OF THE INVENTION

Embodiment 1. 1 is a block diagram showing a schematic configuration of a first embodiment of the present invention. In FIG. 1, 2A and 3A correspond to an ECU 2 and a CPU 3, respectively, and 1, 10 to 15, A, D and Dc are It is similar to the above. Further, the detection processing operation in the normal control mode by the image processing circuit 13 in the ECU 2A and the CPU 3A is the same as described above.

In this case, the CPU 3A uses the image processing circuit 1
Circuit abnormality detecting means 16 for diagnosing presence / absence of abnormality 3
The road abnormality detection means 16 outputs the switching signal E at a predetermined timing (abnormality diagnosis mode) set in advance so that the image processing circuit 13 executes the abnormality detection processing.

Reference numeral 21 denotes the image memory 12 and the image processing circuit 13.
And a first switching input terminal 21a and a second switching input terminal 21b, and a common output terminal 21c, the first switching input terminal 21a of the image memory 12. Common output terminal 21c connected to the output terminal
Is connected to the input terminal of the image processing circuit 13. The switching means 21 responds to the switching signal E from the CPU 3A,
From the first switching input terminal 21a to the second switching input terminal 21
It is designed to switch to b.

Reference numeral 22 is a test pattern memory for storing a predetermined test pattern T generated by the CPU 3A, the output terminal of which is connected to the second switching input terminal 21b of the switching means 21. The test pattern T is composed of a pictorial symbol or the like in which the number of edge pixels is predetermined to a predetermined number.

For example, as the test pattern T, a pattern including a vertical edge component and a horizontal edge component corresponding to the white line 43 and the preceding vehicle 44 (see FIG. 14) is suitable. Further, the test pattern T may be stored in the test pattern memory 22 in advance without being given from the CPU 3A.

The CPU 3A, when executing the abnormality detection processing,
A switching signal E is output to the switching means 21 to switch from the first switching input terminal 21a to the second switching input terminal 21b, and instead of the image data D from the image memory 12, the test from the test pattern memory 22 is performed. The pattern T is selected and input to the image processing circuit 13.

Therefore, the image processing circuit 13 causes the test pattern memory 22 to operate when the abnormality diagnosis mode is switched.
Various kinds of image processing are performed on the image of the test pattern T stored in, and the processing result is input to the CPU 3A. Thereby, the circuit abnormality detecting means 16 in the CPU 3A
On the basis of the binarized data Dc based on the test pattern T, the presence or absence of abnormality of the image processing circuit 13 is diagnosed.

FIG. 2 is a flow chart showing the processing operation of the CPU 3A, and S14 and S15 are the same steps as described above (see FIG. 12). In FIG. 2, CPU3
In order to select the processing operation mode, A first determines whether or not it is in the abnormality diagnosis mode (step S1).

If the normal processing mode (that is, NO)
If so, the white line detection process (step S14) and the preceding vehicle detection process (step S15) described above are executed. If the abnormality diagnosis mode (that is, YES), the circuit abnormality detection process by the circuit abnormality detection means 16 (step). Execute S16).

In step S1, it is determined that the abnormality diagnosis mode is set when the normal control mode is not in operation (when the engine is not in the predetermined operating state, for example) or when the normal control mode is not necessary (when the constant speed control mode is not selected). Then, the process of the abnormality diagnosis mode is selected. Therefore, the circuit abnormality detecting means 16
Includes timing search means for constantly searching for the timing of a predetermined abnormality diagnosis mode, and outputs the switching signal E at this timing to execute the circuit abnormality detection processing.

For example, the circuit abnormality detecting means 16 determines the abnormality diagnosis mode when the constant speed traveling (automatic follow-up) control mode based on the driver's intention is not selected (that is, when the peripheral object detection processing is unnecessary). Then, the switching signal E is output to output the binarized data Dc based on the test pattern T to the CP.
Input to U3A.

Further, even if the constant speed control mode based on the driver's intention is selected, the circuit abnormality detecting means 16 does not actually perform the constant speed control when the driving condition of the vehicle satisfies the following conditions. Since it is not executed, the switching signal E is output by determining the abnormality diagnosis mode, and the image data D based on the test pattern T is input to the CPU 3A.

That is, when the host vehicle is starting from the stopped state, the traveling speed is equal to or lower than a predetermined vehicle speed (for example, 40 km / h corresponding to the lower limit value of the steady vehicle speed), or the steering angle is at a predetermined angle (for example, In the case of 30 ° or more, which corresponds to the upper limit value of steady running, the abnormality diagnosis mode is determined.

Thus, after executing the processing steps S14 and S15 in the normal processing mode or the circuit abnormality detection processing step S16 in the abnormality diagnosis mode, it is determined whether or not an abnormality has occurred in the image processing circuit 13 (step S18). ). If it is determined that no abnormality has occurred (that is, NO), the process returns to step S1 and the above process is repeated.

If it is determined that an abnormality has occurred in the image processing circuit 13 (that is, YES), the ECU 2A
The function of the entire image processing apparatus including the above is stopped to stop the processing operation of the white line position detecting means 14 and the preceding vehicle position detecting means 15 in the normal control mode. Thus
After preventing the erroneous detection information of the white line position and the preceding vehicle position from being reflected in the control, the process of FIG. 2 ends.

FIG. 3 is a flow chart showing a specific processing procedure of the processing executed by the image processing circuit 13 and the circuit abnormality detection processing step S16 executed by the circuit abnormality detecting means 16. S20 and S21 are the same as those described above (see FIG. Thirteen
Refer to the above).

FIGS. 4 to 9 are explanatory views showing data after image processing for the test pattern T. FIG. 4 corresponds to FIG. 17 described above, FIG. 5 corresponds to FIG. 18 (a), and FIG. 7 and FIG. 7 correspond to FIG. 18B, and FIGS.
It corresponds to 8 (c).

In FIG. 4, FIG. 6 and FIG.
40c is a test pattern image corresponding to the imaging frame 40, 40a is an initial test pattern image, and 40b is a test pattern image.
Is an image after edge extraction processing, and 40c is an image after binarization processing. 41A and 41B are the images 40a-.
Scan lines at different vertical positions within 40c.

Further, in FIGS. 5, 7 and 9 showing the luminance data at the horizontal position of each of the images 40a to 40c,
(A) shows the brightness value corresponding to each image data by the scanning line 41A, (b) shows the brightness value corresponding to each image data by the scanning line 41B.

Next, the circuit abnormality detection processing operation according to the first embodiment of the present invention shown in FIG. 1 will be described with reference to FIGS. In this case, the test pattern T
Is composed of a "day" character-shaped pattern as shown in FIG.

First, when the switching means 21 selects the switching input terminal 21b in response to the switching signal E from the CPU 3A, the test pattern T is input to the image processing circuit 13. At this time, the image data Da on each scanning line 41A and 41B for the initial test pattern image 40a is
It becomes like Fig.5 (a) and (b).

Next, referring to FIG. 3, the image processing circuit 13
When the edge extraction processing (step S20) is performed on the initial test pattern image 40a (see FIG. 4), the edge extraction image 40b of the test pattern T is obtained as shown in FIG. The edge image data Db of is as shown in FIGS. 7A and 7B.

When the image processing circuit 13 performs the binarization processing (step S21) on the edge extraction image 40b, the binary image 40c of the test pattern T is obtained as shown in FIG. 8 and each scanning line 41A is obtained. And the binarized data Dc on 41B are as shown in FIGS. 9 (a) and 9 (b).

Next, the circuit abnormality detecting means 1 in the CPU 3A
6 detects the projection value (the number of edge pixels) SUM of the binary image 40c (see FIG. 8) obtained in step 21 (step S50). At this time, the projection value SUM is the binary image 4
The image size is calculated as 640 by the sum of the pixel values of 0c.
× 480 pixels, and the value of each pixel is Gij (i = 1 to 64)
0, j = 1 to 480), it is obtained by the following equation (2).

SUM = ΣΣ (Gij) (2)

However, in the equation (2), the total sum variables i and j are added within the range of i = 1 to 640 and j = 1 to 480. Then, the projection value SUM obtained in step S50 and the preset determination value (predetermined number of pixels) SUMr are compared and determined, and it is determined whether or not there is an abnormality in the image processing circuit 13 based on the matching state between the two.

The determination value SUMr is a value corresponding to the ideal projection value SUM obtained when the test pattern T is input to the normal image processing circuit 13. However, in actuality, the synchronous separation of the CCD camera 1 is not perfect, the test pattern T may be slightly deviated, and the influence of noise may be considered. Therefore, the circuit abnormality detecting means 16
In step S51, the projection value SUM is the judgment value SUM
If the difference from r exceeds the allowable range, it is determined that there is an abnormality.

Here, a specific comparison / determination operation will be described with reference to an example of a simplified binary image 40c as shown in FIG. In FIG. 10, the number i of pixels in the horizontal direction is 25, the number j of pixels in the vertical direction is 20, white pixels indicate a portion having a data value of “0”, and black pixels indicate a portion having a data value of “1”. There is.

In this case, the ideal projection value SUM for the test pattern T is 128 according to equation (2), and this value "128" is used as the judgment value SUMr. Therefore, the permissible range of ± 20% (≈26) of the judgment value SUMr (= 128) is 102 to 154, and the possible value of the projection value SUM when there is no abnormality is as shown in the following formula (3). Represented.

102 <SUM <154 (3)

As is apparent from the equation (3), the projection value SUM of the test pattern T calculated by the equation (2) is equal to or lower than the lower limit value 102 or the upper limit value 15 of the judgment value SUMr.
When the value is 4 or more, the projection value SUM and the judgment value S
It is considered that the UMr does not match, and the image processing circuit 13 is determined to have an abnormality.

As described above, in the abnormality diagnosis mode, the switching means 21 is operated by the switching signal E, the test pattern T is input to the image processing circuit 13 instead of the image data D captured, and based on the test pattern image 40a. Detecting an abnormality in the image processing circuit 13 by obtaining the edge image 40b and the binary image 40c, and comparing the determination value SUMr that can be obtained from the shape of the test pattern T with the actually obtained projection value SUM. You can

Further, the timing searching means in the circuit abnormality detecting means 16 constantly monitors the selected state or operating state of the normal control mode for executing the constant speed operation, and when the normal control mode is not operating (steady state). When it is in a non-operating state or when it is not needed (in the non-selected state of the normal control mode), it is determined to be the abnormality diagnosis mode and the circuit abnormality detection processing can be automatically executed.

Thus, it is possible to automatically detect the occurrence of an abnormality in the image processing circuit 13 not only at the time of startup but also during traveling, and at the time of occurrence of an abnormality, the entire apparatus and the white line position detecting means 14 and It is possible to reliably stop the processing operation of the preceding vehicle position detecting means 15.

Therefore, the image processing circuit 1 in which the abnormality has occurred
No processing is performed by using No. 3, and a highly reliable image processing apparatus can be obtained. Further, since the circuit abnormality detecting means 16 is constantly searching for the abnormality diagnosis mode, the processing time can be shortened and the abnormality diagnosis accuracy can be improved.

Embodiment 2. In the first embodiment, the processing function of each of the detecting means 14 and 15 is automatically stopped when the circuit abnormality is detected. However, the abnormality detection state is displayed as an alarm to prompt the driver to perform a stop operation. Good.

Third Embodiment Further, in the first embodiment, when the normal control is not operated or when it is not necessary is used as the selection condition of the abnormality diagnosis mode in the determination step S1 in FIG. 2, but the present invention is not limited to this condition and detection is performed. It goes without saying that it can be arbitrarily set according to the usage conditions of the result and various required specifications.

Fourth Embodiment Further, in the first embodiment, when the circuit abnormality is detected, the circuit abnormality is determined using the binary image data Dc corresponding to the edges in both the vertical and horizontal directions. Alternatively, the determination may be performed using binary image data corresponding to edges in only one direction.

Embodiment 5. Further, in the above-described first embodiment, the white line 43 and the preceding vehicle 44 are imaged around the vehicle in consideration of the constant speed control mode. Any other peripheral object located in front may be used as the imaging detection target.

[0086]

As described above, according to claim 1 of the present invention, the image pickup means (CCD camera 1) mounted on the vehicle for generating the image signal A for photographing the periphery of the vehicle, and the image signal A. An image processing circuit 13 that extracts edge image data Db from the image data Da based on the image data and performs a binarization process, and a position detection unit that detects a position of a peripheral object of the vehicle based on the binarized data Dc from the image processing circuit 13. 14 and 15,
A test pattern memory 22 in which a predetermined test pattern T is stored, switching means 21 for switching an input signal to the image processing circuit 13 from the image data Da to the test pattern T in response to the control state of the vehicle, and the image processing circuit. 1
Circuit abnormality detection means 16 for detecting an abnormality of the image processing circuit 13 based on the binarized data Dc of the test pattern T through the circuit pattern detection circuit 16 and the circuit abnormality detection means 16 of the edge image data based on the test pattern T. Number of pixels (projection value SU
M) and a predetermined number of pixels serving as an abnormality determination reference value (determination value SU
Since it is determined that the image processing circuit 13 is abnormal when the matching condition between the both is not satisfied, the in-vehicle image processing capable of preventing the output of the erroneously detected data when the circuit is abnormal. There is an effect that the device can be obtained.

According to a second aspect of the present invention, in the first aspect, the circuit abnormality detecting means 16 sets the projection value SUM which is the sum of the pixel values Gij of the edge image data based on the test pattern T to the number of pixels. As the projection value SU
Since M is compared with the determination value SUMr that is the predetermined number of pixels and the projection value SUM deviates from the allowable range based on the determination value SUMr, it is determined that the image processing circuit 13 is abnormal. There is an effect that an in-vehicle image processing device capable of preventing data output can be obtained.

According to claim 3 of the present invention, the circuit abnormality detecting means 16 according to claim 1 or 2 is provided.
When the abnormality of the image processing circuit 13 is determined, the processing functions of the position detecting means 14 and 15 are stopped, so that the output of the erroneous detection data at the time of the circuit abnormality is surely prevented. There is an effect that can be obtained.

According to a fourth aspect of the present invention, in any one of the first to third aspects, the circuit abnormality detecting means 16 searches the timing corresponding to the control state of the vehicle as the abnormality diagnosis mode. When the abnormality diagnosis mode is searched, the switching means 21 includes the timing search means.
Since the switching signal E for operating the above is output, it is possible to obtain the vehicle-mounted image processing apparatus in which the processing time is shortened and the abnormality diagnosis accuracy is improved.

Further, according to claim 5 of the present invention, in any one of claims 1 to 4, the peripheral object detected by the position detecting means 14 and 15 is a white line 43 on the traveling lane of the vehicle. Since at least one of the preceding vehicle 44 and the preceding vehicle 44 is used, it is possible to effectively prevent erroneous data from being output for a peripheral object that is effective for vehicle control, and is particularly suitable for in-vehicle use when the detection result is reflected in the constant speed control mode. The image processing device can be obtained.

According to a sixth aspect of the present invention, in the fifth aspect, there is provided a control mode selecting means for selectively enabling the processing functions of the position detecting means 14 and 15 to set the normal control mode, and timing search. The means determines the abnormality diagnosis mode when the normal control mode of the vehicle is not selected,
Since the processing function of the circuit abnormality detecting means 16 is made effective, there is an effect that a vehicle-mounted image processing apparatus can be obtained in which the processing time is shortened and the abnormality diagnosis accuracy is improved.

According to a seventh aspect of the present invention, in the sixth aspect, the timing searching means is in a control state in which the processing functions of the position detecting means 14 and 15 are disabled when the normal control mode is selected. Since the abnormality diagnosis mode is determined and the processing function of the circuit abnormality detection means 16 is enabled, the vehicle-mounted image processing apparatus capable of effectively searching for the abnormality diagnosis mode even in the normal control mode is provided. There is an effect to be obtained.

According to the eighth aspect of the present invention, in the seventh aspect, the control state in which the processing function of the position detecting means is invalid is set at the time of starting the vehicle. Therefore, even in the normal control mode. There is an effect that an in-vehicle image processing device capable of effectively searching the abnormality diagnosis mode is obtained.

According to the ninth aspect of the present invention, in the seventh aspect, the control state in which the processing function of the position detecting means is disabled is a state in which the traveling speed of the vehicle is equal to or lower than the predetermined vehicle speed. There is an effect that a vehicle-mounted image processing apparatus can be obtained that can effectively search for an abnormality diagnosis mode even during the mode.

According to the tenth aspect of the present invention, in the seventh aspect, the control state in which the processing function of the position detecting means is disabled is the state in which the steering angle of the vehicle is equal to or more than the predetermined angle. There is an effect that a vehicle-mounted image processing apparatus can be obtained that can effectively search for an abnormality diagnosis mode even during the mode.

[Brief description of drawings]

FIG. 1 is a block diagram schematically showing a basic configuration of a first embodiment of the present invention.

FIG. 2 is a flowchart showing a mode switching processing operation of the circuit abnormality detecting means according to the first embodiment of the present invention.

FIG. 3 is a flowchart showing a circuit abnormality detecting operation of the circuit abnormality detecting means according to the first embodiment of the present invention.

FIG. 4 is an explanatory diagram showing a test pattern image according to the first embodiment of the present invention.

FIG. 5 is an explanatory diagram showing data levels on each scanning line of the test pattern image according to the first embodiment of the present invention.

6 is an explanatory diagram showing an edge extraction image obtained by performing edge extraction processing on the test pattern image in FIG. 4. FIG.

7 is an explanatory diagram showing data levels on each scanning line of the edge extraction image in FIG.

FIG. 8 is an explanatory diagram showing a binary image obtained by performing binarization processing on the edge extraction image in FIG.

9 is an explanatory diagram showing data levels on each scanning line of the binary image in FIG.

FIG. 10 is an explanatory diagram showing an example of a binary image for specifically explaining the determination condition of the circuit abnormality detecting means according to the first embodiment of the present invention.

FIG. 11 is a block diagram showing a basic configuration example of a conventional vehicle-mounted image processing apparatus.

FIG. 12 is a flowchart showing a peripheral object detection processing operation by a conventional vehicle-mounted image processing apparatus.

FIG. 13 is a flowchart showing a white line detection processing operation by a conventional vehicle-mounted image processing apparatus.

FIG. 14 is an explanatory diagram showing an example of an image for one frame captured while the vehicle is traveling.

FIG. 15 is an explanatory diagram showing a process when white line position detection processing is performed on the image example in FIG. 14;

FIG. 16 is a flowchart showing a preceding vehicle detection processing operation by a conventional vehicle-mounted image processing apparatus.

FIG. 17 is an explanatory diagram showing an example of an image for one frame captured while the vehicle is traveling.

FIG. 18 is an explanatory diagram showing a process when a preceding vehicle position detection process is performed on the image example in FIG. 17.

[Explanation of symbols]

1 CCD camera (imaging means), 2A ECU, 3A
CPU, 12 image memory, 13 image processing circuit, 14
White line position detecting means, 15 Leading vehicle position detecting means, 16
Circuit abnormality detecting means, 21 switching means, 22 test pattern memory, 40 imaging frame, 40a test pattern image, 40b edge extraction image, 40c binary image, 41A, 41B scanning line, 43 white line, 44 preceding vehicle, A image signal, D, Da image data, Db edge extraction image data, Dc binarized data, E switching signal,
SUM projection value (number of images), SUMr determination value (predetermined number of images), T test pattern, S1 step of determining abnormality diagnosis mode, S14 white line detection processing step, S
15 preceding vehicle detection processing step, S16 circuit abnormality detection processing step, S18 circuit abnormality determination step, S20
Edge extraction processing step, S21 binarization processing step, S50 detecting projection value, S51 comparing projection value with judgment value.

Front page continuation (51) Int.Cl. ⁷ identification code FI H04N 7/18 H04N 7/18 J (58) Fields investigated (Int.Cl. ⁷ , DB name) G06T 7/60 G01B 11/00 G06T 1 / 00 G08B 21/00 G08G 1/16 H04N 7/18

Claims (10)

(57) [Claims] 1. An image pickup means mounted on a vehicle for generating an image signal of the periphery of the vehicle, and an image processing circuit for extracting edge image data from image data based on the image signal and performing binarization processing. Position detecting means for detecting a position of a peripheral object of the vehicle based on the binarized data from the image processing circuit, a test pattern memory storing a predetermined test pattern, and a response to a control state of the vehicle. Switching means for switching the input signal to the image processing circuit from the image data to the test pattern, and detecting an abnormality of the image processing circuit based on the binarized data of the test pattern via the image processing circuit. Circuit abnormality detection means, the circuit abnormality detection means, the number of pixels of the edge image data based on the test pattern, and the abnormality determination criteria A vehicle-mounted image processing device, which compares a predetermined number of pixels as a value and determines that the image processing circuit is abnormal when the matching condition between the two is not satisfied. 2. The circuit abnormality detecting means obtains, as the number of pixels, a projection value that is a sum of pixel values of edge image data based on the test pattern, and sets the projection value as a determination value that is the predetermined number of pixels. The vehicle-mounted image processing device according to claim 1, wherein the abnormality is detected in the image processing circuit when the projection value deviates from an allowable range based on the determination value. 3. The vehicle-mounted system according to claim 1 or 2, wherein the circuit abnormality detecting means stops the processing function of the position detecting means when the abnormality of the image processing circuit is determined. Image processing device. 4. The circuit abnormality detection means includes a timing search means for searching a timing according to a control state of the vehicle as an abnormality diagnosis mode, and operates the switching means when the abnormality diagnosis mode is searched. A vehicle-mounted image processing device according to any one of claims 1 to 3, wherein a switching signal for outputting the signal is output. 5. The surrounding object detected by the position detecting means is at least one of a white line on a lane of travel of the vehicle and a preceding vehicle, according to any one of claims 1 to 4. The in-vehicle image processing device according to item 1. 6. A control mode selecting unit that selectively enables a processing function of the position detecting unit to set a normal control mode, and the timing searching unit diagnoses an abnormality when the normal control mode of the vehicle is not selected. The vehicle-mounted image processing apparatus according to claim 5, wherein the mode is determined and the processing function of the circuit abnormality detection means is enabled. 7. The timing search means determines the abnormality diagnosis mode in a control state in which the processing function of the position detection means is disabled when the normal control mode is selected, and the processing of the circuit abnormality detection means is performed. The vehicle-mounted image processing device according to claim 6, wherein the function is enabled. 8. The vehicle-mounted image processing apparatus according to claim 7, wherein the control state in which the processing function of the position detecting means is disabled is when the vehicle is starting. 9. The vehicle-mounted image processing apparatus according to claim 7, wherein the control state in which the processing function of the position detecting means is disabled is a state in which the traveling speed of the vehicle is equal to or lower than a predetermined vehicle speed. 10. The vehicle-mounted image processing apparatus according to claim 7, wherein the control state in which the processing function of the position detecting means is disabled is a state in which the steering angle of the vehicle is equal to or larger than a predetermined angle.