Cat-eye effect target detection method of a collaborative filtering reflector device

Abstract

The present application belongs to the technical field of photoelectric countermeasure, and relates to a detection technology for a hidden camera device, and specifically provides a cat-eye effect target detection method for filtering a reflective device, to solve the problem of low recognition accuracy caused by misjudgment of the reflective device in the existing cat-eye target detection method. Through technical means such as image edge detection, image brightness uniformization processing, image arithmetic operation, image pixel value comparison operation and binary graph contour area determination, the present application deletes edge noise information, noise information caused by uneven brightness, static background information, dynamic background information and reflective device area information of the original image, to realize cat-eye target detection for filtering the reflective device, effectively improve the recognition accuracy of the cat-eye target, and further realize anti-detection of the hidden camera device and eliminate potential information leakage danger.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of photoelectric countermeasure, and relates to a detection technology for a hidden camera device, and particularly provides a cat-eye effect target detection method for filtering a reflective device. BACKGROUND

[0002] With the wide application of pinhole cameras, miniature cameras and other photoelectric detection equipment, various activities are facing more and more serious information leakage risks, and the leakage of confidential information will cause great losses in military, social and commercial fields. The most common method for detecting a hidden camera device is a laser active detection method, which is mainly based on the characteristics of cat-eye effect. After the laser actively irradiates the hidden camera device, the lens of the hidden camera device will reflect most of the light. Generally, the device obtains an image containing a cat-eye target after differentiating the active and passive images. The common problem of such devices is that when there is a reflective device, the reflective device will be misjudged as a hidden camera device by the system, greatly reducing the recognition accuracy of the system. SUMMARY

[0003] The present application aims to solve the problem of low recognition accuracy caused by misjudgment of the existing cat-eye target detection method due to reflective devices, and provides a cat-eye effect target detection method for filtering a reflective device. The present application deletes edge noise information, noise information caused by uneven brightness, static background information, dynamic background information and reflective device area information of the original image by using image edge detection, image brightness uniformization processing, image arithmetic operation, image pixel value comparison operation and binary graph contour area determination, so as to realize cat-eye target detection for filtering a reflective device, effectively improve the recognition accuracy of the cat-eye target, and further realize anti-detection of a hidden camera device and eliminate potential information leakage risks.

[0004] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows:

[0005] A cat-eye effect target detection method for filtering a reflective device, characterized in that it comprises the following steps:

[0006] S1. Collecting a pair of active and passive images by a laser active detection device;

[0007] S2. Performing the same gray scale processing on the active and passive images and removing the edge part;

[0008] S3. Performing brightness uniformization processing on the active image obtained in step S2;

[0009] S4. Performing difference operation on the active image obtained in step S3 and the passive image obtained in step S2 to obtain a difference image Diff;

[0010] S5. The active image and the passive image of step S2 are respectively subjected to the same binarization processing, and the binarized images obtained by the processing are added to obtain a binarized image Binary;

[0011] S6. The binarized image Binary is subjected to contour detection, the contour area of the region with pixel value of 255 in the binarized image Binary is detected, and the region with contour area greater than a threshold T2 is determined as a reflective device, and a corresponding reflective device information image Reflect is obtained;

[0012] S7. The difference image Diff is subtracted from the reflective device information image Reflect to obtain an image Df after filtering the reflective device;

[0013] S8. The pixel value of the region with pixel value lower than a threshold T3 in the image Df of step S7 is set to 0;

[0014] S9. The connected region of the non-0 pixel value in the image of step S8 is detected, and is determined as a cat eye target region, and the cat eye target is marked in the original active image of step S1.

[0015] Further, in step S2, the specific process of removing the edge part is as follows: the edges of the image are removed by 2% respectively.

[0016] Further, in step S3, the specific process of the brightness uniformization processing is as follows: each pixel point of the active image is subtracted by the average value of all pixel points of the image.

[0017] Further, in step S5, the specific process of the binarization processing is as follows: for the active image and the passive image of step S2, the threshold T1 is set to 1.5 times of the average pixel value of the image, the pixel value lower than the threshold is set to 0, and the pixel value higher than the threshold is set to 255, to obtain the binarized active image and the binarized passive image.

[0018] Further, in step S6, the threshold T2 is 2% of the image area.

[0019] Further, in step S8, the threshold T3 is 3 / 4 times of the highest pixel value of the image.

[0020] Based on the above technical scheme, the beneficial effects of the present application are as follows:

[0021] The application provides a cat-eye effect target detection method of a collaborative filtering reflection equipment, which deletes edge noise information, noise information caused by uneven brightness, static background information, dynamic background information and reflection equipment region information of an original image through image edge detection, image brightness uniformization processing, image arithmetic operation, image pixel value comparison operation and binary image contour region determination, so as to realize cat-eye target detection of the collaborative filtering reflection equipment, effectively improve the recognition accuracy of the cat-eye target, and further realize anti-detection of a hidden camera and eliminate potential information leakage danger. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 It is a flowchart of the cat-eye effect target detection method of the collaborative filtering reflection equipment in the application.

[0023] Figure 2 It is the active image and the passive image after step S1 in the embodiment of the application.

[0024] Figure 3 It is the active image and the passive image after step S2 in the embodiment of the application.

[0025] Figure 4 It is the active image after step S3 in the embodiment of the application.

[0026] Figure 5 It is the difference image Diff after step S4 in the embodiment of the application.

[0027] Figure 6 It is the binary image of the active image and the passive image after step S5 in the embodiment of the application.

[0028] Figure 7 It is the binary image Binary after step S5 in the embodiment of the application.

[0029] Figure 8 It is the reflection equipment information image Reflect after step S6 in the embodiment of the application.

[0030] Figure 9 It is the image Df after filtering the reflection equipment after step S7 in the embodiment of the application.

[0031] Figure 10 It is the image after filtering the reflection equipment after step S8 in the embodiment of the application.

[0032] Figure 11 It is the cat-eye target detection result after step S9 in the embodiment of the application. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical scheme and technical effects of the present application more clear, the present application is further described in detail below with reference to the drawings and examples.

[0034] The embodiment provides a cat-eye effect target detection method for collaborative filtering of a reflective device, which is used for realizing filtering of the reflective device, and thus improving the recognition accuracy of cat-eye target detection. The flow of the cat-eye target detection method is shown in Figure 1 and specifically includes the following steps.

[0035] S1. Collecting a pair of active images and passive images through a laser active detection device, wherein the active image refers to an image of a cat-eye target detection area under laser irradiation, and the passive image refers to an original image of the cat-eye target detection area (without laser irradiation); as shown in Figure 2 A1 is the active image, A2 is the passive image, and the image size is 400x400 (lengthxwidth).

[0036] S2. Performing the same gray-scale processing on the active image and the passive image and eliminating the edge part. The gray-scale processing can convert the 3-channel RGB image into a 1-channel gray-scale image, greatly reducing the data storage amount of the image and facilitating the shortening of the time for subsequent cat-eye target detection in the image. The edge part of the image generally contains a large amount of noise information, so the edge part is eliminated.

[0037] In the embodiment, for the 400x400 active image and passive image, 10 pixel points are respectively eliminated from each edge (both ends of length and width) of the image, and the image size after processing is 384x384 (lengthxwidth). As shown in Figure 3 B1 is the active image, and B2 is the passive image.

[0038] S3. Performing brightness uniformization processing on the active image B1 obtained in step S2. Because the light intensity of some parts of the image is different under natural light, some parts are seriously exposed when the laser is turned on, and after difference, it is extremely likely to mistake the part with extremely strong light as a "cat-eye" target, which is extremely unfavorable for subsequent difference operation. Therefore, each pixel point of the active image is subtracted by the average value of all pixel points of the image, that is, the brightness uniformization processing is realized, the exposed area in the image is darkened, which is favorable for difference operation and improves the detection accuracy of the algorithm. The active image after processing is shown in Figure 4 C1 is the active image.

[0039] S4. Performing difference operation on the active image C1 obtained in step S3 and the passive image B2 obtained in step S2 to obtain a difference image Diff: Diff=C1-B2, as shown in Figure 5As shown, the difference image Diff after the difference operation removes the static background noise information from the original image, but it still contains dynamic background noise information, reflective device information, and cat eye target information that need to be detected.

[0040] S5. Perform the same binarization process on the active image B1 and passive image B2 obtained in step S2. The threshold T1 for binarization is set to 1.5 times the average pixel value of the image. Pixel values ​​below the threshold are set to 0, and pixel values ​​above the threshold are set to 255, resulting in the binarized image Binary1 corresponding to the active image and the binarized image Binary2 corresponding to the passive image. Figure 6 As shown; then, the binarized image Binary1 is added to the binarized image Binary2 (the corresponding pixel values ​​are added) to obtain the binarized image Binary. The binarized image Binary removes information from B1 and B2 that is less than 1.5 times the average pixel value, as shown. Figure 7 As shown;

[0041] S6. Perform contour detection on the binarized image Binary from step S5. Detect the contour area of ​​regions with a pixel value of 255 in the binarized image Binary, and identify regions with contour areas greater than a threshold T2 as reflective devices, thus obtaining the reflective device information image Reflect. Figure 8 As shown; the threshold T2 is 2% of the image area (384×384);

[0042] S7. Subtract the difference image Diff obtained in step S4 from the reflective device information image Reflect obtained in step S6 to obtain the image Df after filtering the reflective device: Df = Diff - Reflect, as shown. Figure 9 As shown;

[0043] S8. Set the pixel values ​​of regions in image Df from step S7 that have pixel values ​​lower than the threshold T3 to 0, and exclude them from the cat's eye target region. Figure 10 As shown; the threshold T3 is 3 / 4 times the highest pixel value of the image;

[0044] S9. Detect connected regions with non-zero pixel values ​​in the image from step S8 and identify them as cat-eye target regions. Mark the cat-eye target in the original active image from step S1, such as... Figure 11 As shown.

[0045] In this embodiment, a laser active detection device is constructed using a base plate, a PyAIOpenMV4 development board, and a semiconductor near-infrared dot matrix laser (with a power of 2W). Based on the above-mentioned method for detecting surreptitious cameras using collaborative filtering reflective devices, the location and labeling of the "cat's eye" target are ultimately achieved.

[0046] The above merely provides the specific implementation of the present application, any feature disclosed in the specification can be replaced by other equivalent or similar purpose alternative features unless specifically described, and all features disclosed or all steps in the method or process can be combined in any manner except for mutually exclusive features and / or steps.

Claims

1. A method for detecting cat's-eye effect targets in a collaborative filtering reflective device, characterized in that, Includes the following steps: S1. Acquire a pair of active and passive images using a laser active detection device; S2. Perform the same grayscale processing on both the active and passive images, and remove edge parts; S3. Perform brightness uniformization processing on the active image obtained in step S2; S4. Perform a difference operation on the active image obtained in step S3 and the passive image obtained in step S2 to obtain the difference image Diff; S5. Perform the same binarization process on the active image and the passive image obtained in step S2, and add the binarized images to obtain the binarized image Binary; S6. Perform contour detection on the binarized image Binary, detect the contour area of ​​the region with a pixel value of 255 in the binarized image Binary, and determine the region with a contour area greater than the threshold T2 as a reflective device, and obtain the corresponding reflective device information image Reflect; S7. Subtract the difference image Diff from the reflector information image Reflect to obtain the image Df after filtering the reflector; S8. Set the pixel values ​​of regions in image Df whose pixel values ​​are lower than the threshold T3 after step S7 to 0; S9. Detect the connected regions with non-zero pixel values ​​in the image after step S8, and determine them as the cat eye target region. Mark the cat eye target in the original active image of step S1.

2. The cat-eye target detection method of the collaborative filtering reflective device according to claim 1, characterized in that, In step S2, the specific process of removing edge parts is as follows: remove each edge of the image at a ratio of 2%.

3. The cat-eye target detection method of the collaborative filtering reflective device according to claim 1, characterized in that, In step S3, the specific process of brightness uniformization is as follows: subtract the average value of all pixels in the image from each pixel of the active image.

4. The cat-eye target detection method of the collaborative filtering reflective device according to claim 1, characterized in that, In step S5, the specific process of binarization is as follows: For the active image and passive image obtained in step S2, the threshold T1 is set to 1.5 times the average pixel value of the image. Pixel values ​​below the threshold are set to 0, and pixel values ​​above the threshold are set to 255, thus obtaining the binarized active image and passive image.

5. The cat-eye target detection method of the collaborative filtering reflective device according to claim 1, characterized in that, In step S6, the threshold T2 is 2% of the image area.

6. The cat-eye target detection method of the collaborative filtering reflective device according to claim 1, characterized in that, In step S8, the threshold T3 is 3 / 4 times the highest pixel value of the image.

Images