Target recognition device and method based on industrial camera image

Abstract

The invention discloses a target recognition device and method based on industrial camera images, and relates to the technical field of industrial cameras. The target recognition device based on the industrial camera image comprises an industrial camera base, a lens and an image light supplementing structure are fixedly installed on the industrial camera base, the image light supplementing structure is located on the industrial camera base and comprises a sliding rail, a gear ring, a gear, a driving motor and a sliding base, and the sliding rail is fixedly connected to the side, close to the lens, of the industrial camera base; the sliding seat is slidably connected to the sliding rail, the gear ring is fixedly connected to the outer surface of the sliding rail in a sleeving mode, through cooperation of the sliding rail, the gear ring, the gear, the sliding seat, the first rotating rod, the second rotating rod and the light supplementing lamp, the illumination angle of the light supplementing lamp can be adjusted through the first rotating rod, the second rotating rod and the light supplementing lamp according to needs, and the illumination angle of the light supplementing lamp can be adjusted through meshing connection of the gear and the gear ring. Therefore, the sliding seat can drive the light supplement lamp to supplement light to the shadow of the image, and the recognition quality of the lens image is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial cameras, and in particular to a target recognition device and method based on industrial camera images. Background Art

[0002] With the rapid development of industrial automation and machine vision technology, image target recognition devices based on industrial cameras have been widely used in intelligent manufacturing, quality inspection, automatic sorting and other fields. Such devices can quickly and accurately identify target objects by collecting and analyzing image data, thereby greatly improving production efficiency and product quality.

[0003] When existing industrial cameras recognize images, the lighting conditions have a significant impact on the quality of image acquisition and the accuracy of target recognition. Especially in complex and changeable industrial environments, problems such as uneven lighting and shadow interference often lead to a decrease in image quality, which in turn affects the effect of target recognition. The current fill light device is usually fixed in a certain position and cannot flexibly adjust the fill light angle according to the position and posture of the target object, resulting in insufficient lighting in some areas. Therefore, we propose a target recognition device and method based on industrial camera images. Summary of the invention

[0004] The purpose of the present invention is to solve at least one of the technical problems existing in the prior art, and to provide a target recognition device and method based on industrial camera images, which can solve the problem that light conditions have an important impact on the quality of image acquisition and the accuracy of target recognition. Especially in complex and changeable industrial environments, problems such as uneven lighting and shadow interference often lead to a decrease in image quality, which in turn affects the effect of target recognition. The current fill light device is usually fixedly installed in a certain position, and the fill light angle cannot be flexibly adjusted according to the position and posture of the target object, resulting in insufficient light in some areas.

[0005] To achieve the above object, the present invention provides the following technical solution: a target recognition device based on industrial camera images, comprising:

[0006] Industrial camera holder, the industrial camera holder is fixed with a lens;

[0007] Image fill light structure, the image fill light structure is located on the industrial camera base;

[0008] The image fill light structure includes a slide rail, a gear ring, a gear, a drive motor and a sliding seat. The slide rail is fixedly connected to the side of the industrial camera seat close to the lens, the sliding seat is slidably connected to the slide rail, the gear ring is fixedly sleeved on the outer surface of the slide rail, the drive motor is fixedly installed on the sliding seat, the output end of the drive motor is fixedly connected to the gear, and the gear is meshingly connected to the gear ring.

[0009] Preferably, the image fill light structure also includes three rotating motors, a first rotating rod, a second rotating rod and a fill light. The first rotating rod is rotatably connected to the side of the sliding seat away from the sliding rail, the second rotating rod is rotatably connected to the inside of the first rotating rod, and the fill light is rotatably connected to the inside of the second rotating rod. The three rotating motors are respectively fixedly installed at the rotation points of the first rotating rod, the second rotating rod and the fill light, and the output ends of the three rotating motors are respectively fixedly connected to the corresponding first rotating rod, the second rotating rod and the fill light.

[0010] Preferably, a protective cover is fixedly sleeved on the outer surface of the slide rail, a sliding groove is provided on a side of the protective cover close to the lens, and the output end of the drive motor is located in the sliding groove.

[0011] Preferably, the interior of the industrial camera holder is respectively provided with a PFGA board, a back cover, a front cover, a sensor board, a center hole and an interface board, the sensor board is mounted on one side of the front cover, the center hole is fixedly mounted on the side of the sensor board away from the front cover, the PFGA board is fixedly mounted on the side of the center hole away from the sensor board, the interface board is fixedly mounted on the side of the PFGA board away from the center hole, and the back cover is fixedly mounted on the side of the interface board away from the PFGA board.

[0012] Preferably, the interiors of the slide rail and the slide seat are both "T"-shaped structures.

[0013] Methods for target recognition based on industrial camera images include:

[0014] S1: Use the built-in sensor of the industrial camera to capture images of the target object. These images contain rich visual information and are the basis for subsequent processing;

[0015] S2: Convert the color image to a grayscale image to reduce the processing complexity, improve the processing speed, and remove the noise interference in the image. For example, use Gaussian filtering, median filtering and other methods to adjust the contrast, brightness and other parameters to make the target features in the image more prominent, which is convenient for subsequent processing. Convert the image to black and white to facilitate subsequent feature extraction and segmentation.

[0016] S3: Segment the image into multiple regions according to certain conditions, usually based on features such as color and texture, detect edge information in the image, use edge detection algorithms to extract target boundaries, detect connected areas in the image, combine prior knowledge of the target for segmentation, and use deep learning models to automatically learn image features to achieve more complex segmentation tasks;

[0017] S4: extract information related to target features, such as corner points, contours, textures, etc., from the segmented target area, process the extracted information to obtain relevant feature vectors, which will be used for subsequent target recognition;

[0018] S5: Compare the extracted feature vector with the preset template, determine the target category by similarity, use the trained neural network model to automatically learn and identify the target features, perform error correction, typesetting and formatting on the recognition results, and improve the recognition accuracy and readability;

[0019] S6: Feedback the recognition results to the robot motion controller or other related systems to guide subsequent operations or decisions;

[0020] S7: The fill light is enabled to perform fill light processing on the shadow area recognized by the lens.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1. The target recognition device based on industrial camera images, through the cooperation of the slide rail, the gear ring, the gear, the sliding seat, the first rotating rod, the second rotating rod and the fill light, enables the first rotating rod, the second rotating rod and the fill light to adjust the lighting angle of the fill light as needed, and under the meshing connection of the gear and the gear ring, the sliding seat can drive the fill light to fill in the shadow of the image, thereby improving the recognition quality of the lens image.

[0023] 2. The target recognition device based on industrial camera images, through the provision of a protective cover, enables the protective cover to protect the gears and gear rings, thereby improving the practicality of the fill light device. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention is further described below in conjunction with the accompanying drawings and embodiments:

[0025] Figure 1 It is a schematic diagram of the three-dimensional structure of the present invention;

[0026] Figure 2 It is a schematic diagram of the slide rail structure of the present invention;

[0027] Figure 3 It is a schematic diagram of the cross-sectional structure of the slide rail of the present invention;

[0028] Figure 4 It is a schematic diagram of the internal parts structure of the industrial camera holder of the present invention.

[0029] Figure numerals: 1. Industrial camera holder; 2. Lens; 3. Slide rail; 4. Protective cover; 5. Driving motor; 6. Second rotating rod; 7. Fill light; 8. Rotating motor; 9. PFGA board; 10. Back cover; 11. Gear ring; 12. Gear; 13. Sliding seat; 14. First rotating rod; 15. Front cover; 16. Sensor board; 17. Middle pass; 18. Interface board. DETAILED DESCRIPTION

[0030] This section will describe in detail the specific embodiments of the present invention. The preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the accompanying drawings is to supplement the description of the text part of the specification with graphics, so that people can intuitively and vividly understand each technical feature and the overall technical solution of the present invention, but it cannot be understood as a limitation on the scope of protection of the present invention.

[0031] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0032] In the description of the present invention, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0033] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0034] See also Figure 1-4 The present invention provides a technical solution: a target recognition device based on industrial camera images, comprising:

[0035] An industrial camera holder 1, wherein a lens 2 is fixedly mounted on the industrial camera holder 1;

[0036] An image fill light structure, the image fill light structure is located on the industrial camera holder 1;

[0037] The image fill light structure includes a slide rail 3, a gear ring 11, a gear 12, a drive motor 5 and a sliding seat 13. The slide rail 3 is fixedly connected to the side of the industrial camera seat 1 close to the lens 2. The sliding seat 13 is slidably connected to the slide rail 3. The gear ring 11 is fixedly sleeved on the outer surface of the slide rail 3. The drive motor 5 is fixedly installed on the sliding seat 13. The output end of the drive motor 5 is fixedly connected to the gear 12. The gear 12 is meshed with the gear ring 11. The interiors of the slide rail 3 and the sliding seat 13 are both "T"-shaped structures.

[0038] The image fill light structure also includes three rotating motors 8, a first rotating rod 14, a second rotating rod 6 and a fill light 7. The first rotating rod 14 is rotatably connected to the side of the sliding seat 13 away from the sliding rail 3, the second rotating rod 6 is rotatably connected to the inside of the first rotating rod 14, and the fill light 7 is rotatably connected to the inside of the second rotating rod 6. The three rotating motors 8 are respectively fixedly installed at the rotation points of the first rotating rod 14, the second rotating rod 6 and the fill light 7, and the output ends of the three rotating motors 8 are respectively fixedly connected to the corresponding first rotating rod 14, the second rotating rod 6 and the fill light 7.

[0039] A protective cover 4 is fixedly sleeved on the outer surface of the slide rail 3 . A sliding groove is formed on a side of the protective cover 4 close to the lens 2 . The output end of the driving motor 5 is located in the sliding groove.

[0040] The interior of the industrial camera holder 1 is respectively provided with a PFGA board 9, a rear cover 10, a front cover 15, a sensor board 16, a central passage 17 and an interface board 18. The sensor board 16 is installed on one side of the front cover 15, the central passage 17 is fixedly installed on the side of the sensor board 16 away from the front cover 15, the PFGA board 9 is fixedly installed on the side of the central passage 17 away from the sensor board 16, the interface board 18 is fixedly installed on the side of the PFGA board 9 away from the central passage 17, and the rear cover 10 is fixedly installed on the side of the interface board 18 away from the PFGA board 9.

[0041] Methods for target recognition based on industrial camera images include:

[0042] S1: Use the built-in sensor of the industrial camera to capture images of the target object. These images contain rich visual information and are the basis for subsequent processing;

[0043] S2: Convert the color image to a grayscale image to reduce the processing complexity, improve the processing speed, and remove the noise interference in the image. For example, use Gaussian filtering, median filtering and other methods to adjust the contrast, brightness and other parameters to make the target features in the image more prominent, which is convenient for subsequent processing. Convert the image to black and white to facilitate subsequent feature extraction and segmentation.

[0044] S3: Segment the image into multiple regions according to certain conditions, usually based on features such as color and texture, detect edge information in the image, use edge detection algorithms to extract target boundaries, detect connected areas in the image, combine prior knowledge of the target for segmentation, and use deep learning models to automatically learn image features to achieve more complex segmentation tasks;

[0045] S4: extract information related to target features, such as corner points, contours, textures, etc., from the segmented target area, process the extracted information to obtain relevant feature vectors, which will be used for subsequent target recognition;

[0046] S5: Compare the extracted feature vector with the preset template, determine the target category by similarity, use the trained neural network model to automatically learn and identify the target features, perform error correction, typesetting and formatting on the recognition results, and improve the recognition accuracy and readability;

[0047] S6: Feedback the recognition results to the robot motion controller or other related systems to guide subsequent operations or decisions;

[0048] S7: enabling the fill light 7 to perform fill light processing on the shadow area recognized by the lens 2.

[0049] Furthermore, when the device is used, the drive motor 5 is started by connecting an external power source, and the drive motor 5 drives the gear 12 to rotate. Under the meshing connection between the gear 12 and the gear ring 11, the gear 12 can drive the sliding seat 13 to slide on the slide rail 3 under the meshing of the gear ring 11, so as to facilitate adjustment according to the image shadow recognized by the lens 2. The three rotating motors 8 are started by connecting an external power source, so that the three rotating motors 8 respectively drive the first rotating rod 14, the second rotating rod 6 and the fill light 7 to rotate as needed, thereby facilitating the adjustment of the lighting angle of the fill light 7, so as to perform fill light processing on the shadow of the image.

[0050] Through the cooperation of the slide rail 3, the gear ring 11, the gear 12, the sliding seat 13, the first rotating rod 14, the second rotating rod 6 and the fill light 7, the first rotating rod 14, the second rotating rod 6 and the fill light 7 can adjust the lighting angle of the fill light 7 as needed. Under the meshing connection of the gear 12 and the gear ring 11, the sliding seat 13 can drive the fill light 7 to fill in the shadow of the image, thereby improving the recognition quality of the image of the lens 2.

[0051] By providing the protective cover 4, the protective cover 4 can protect the gear 12 and the gear ring 11, thereby improving the practicability of the fill light device.

[0052] Working principle: by connecting an external power source to start the driving motor 5, the driving motor 5 will drive the gear 12 to rotate, and under the meshing connection between the gear 12 and the gear ring 11, the gear 12 can drive the sliding seat 13 to slide on the slide rail 3 under the meshing of the gear ring 11, so as to facilitate adjustment according to the image shadow recognized by the lens 2, and by connecting the external power source to start the three rotating motors 8, the three rotating motors 8 respectively drive the first rotating rod 14, the second rotating rod 6 and the fill light 7 to rotate as needed, so as to facilitate the adjustment of the lighting angle of the fill light 7, so as to perform fill light processing on the shadow of the image.

[0053] The embodiments of the present invention are described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the knowledge scope of ordinary technicians in the technical field without departing from the purpose of the present invention.

Claims

1. A target recognition device based on industrial camera images, characterized in that: include: An industrial camera holder (1), wherein a lens (2) is fixedly mounted on the industrial camera holder (1); An image fill-in light structure, the image fill-in light structure is located on the industrial camera base (1); The image fill light structure comprises a slide rail (3), a gear ring (11), a gear (12), a drive motor (5) and a slide seat (13); the slide rail (3) is fixedly connected to a side of the industrial camera seat (1) close to the lens (2), and the slide seat (13) is slidably connected to the slide rail (3); The gear ring (11) is fixedly sleeved on the outer surface of the slide rail (3), the drive motor (5) is fixedly installed on the slide seat (13), the output end of the drive motor (5) is fixedly connected to the gear (12), and the gear (12) is meshingly connected to the gear ring (11).

2. The target recognition device based on industrial camera images according to claim 1 is characterized in that: The image fill light structure further comprises three rotating motors (8), a first rotating rod (14), a second rotating rod (6) and a fill light (7); the first rotating rod (14) is rotatably connected to a side of the sliding seat (13) away from the sliding rail (3); The second rotating rod (6) is rotatably connected to the inside of the first rotating rod (14), the fill light (7) is rotatably connected to the inside of the second rotating rod (6), and three rotating motors (8) are respectively fixedly installed at the rotating positions of the first rotating rod (14), the second rotating rod (6) and the fill light (7); The output ends of the three rotating motors (8) are respectively fixedly connected to the corresponding first rotating rod (14), the second rotating rod (6) and the fill light (7).

3. The target recognition device based on industrial camera images according to claim 1 is characterized in that: A protective cover (4) is fixedly sleeved on the outer surface of the slide rail (3); a sliding groove is provided on a side of the protective cover (4) close to the lens (2); and an output end of the drive motor (5) is located in the sliding groove.

4. The target recognition device based on industrial camera images according to claim 1 is characterized in that: The industrial camera seat (1) is provided with a PFGA board (9), a rear cover (10), a front cover (15), a sensor board (16), a central passage (17) and an interface board (18) respectively inside. The sensor board (16) is mounted on one side of the front cover (15), the central passage (17) is fixedly mounted on a side of the sensor board (16) away from the front cover (15), the PFGA board (9) is fixedly mounted on a side of the central passage (17) away from the sensor board (16), the interface board (18) is fixedly mounted on a side of the PFGA board (9) away from the central passage (17), and the rear cover (10) is fixedly mounted on a side of the interface board (18) away from the PFGA board (9).

5. The target recognition device based on industrial camera images according to claim 1 is characterized in that: The interiors of the slide rail (3) and the slide seat (13) are both T-shaped structures.

6. A method for target recognition based on industrial camera images, characterized in that: include S1: Use the built-in sensor of the industrial camera to capture images of the target object. These images contain rich visual information and are the basis for subsequent processing; S2: Convert the color image to a grayscale image to reduce the processing complexity, improve the processing speed, and remove the noise interference in the image. For example, use Gaussian filtering, median filtering and other methods to adjust the contrast, brightness and other parameters to make the target features in the image more prominent, which is convenient for subsequent processing. Convert the image to black and white to facilitate subsequent feature extraction and segmentation. S3: Segment the image into multiple regions according to certain conditions, usually based on features such as color and texture, detect edge information in the image, use edge detection algorithms to extract target boundaries, detect connected areas in the image, combine prior knowledge of the target for segmentation, and use deep learning models to automatically learn image features to achieve more complex segmentation tasks; S4: extract information related to target features, such as corner points, contours, textures, etc., from the segmented target area, process the extracted information to obtain relevant feature vectors, which will be used for subsequent target recognition; S5: Compare the extracted feature vector with the preset template, determine the target category by similarity, use the trained neural network model to automatically learn and identify the target features, perform error correction, typesetting and formatting on the recognition results, and improve the recognition accuracy and readability; S6: Feedback the recognition results to the robot motion controller or other related systems to guide subsequent operations or decisions; S7: enabling the fill light 7 to perform fill light processing on the shadow area recognized by the lens 2.

Images