A motion recognition system based on optical marking symbols

Abstract

This invention provides a motion recognition system based on optical marking symbols, belonging to the field of recognition and marking systems. It includes a recognition system and a main control processor. The recognition system includes a camera and a recognition object. The camera captures the recognition object, which is coated with a paint layer forming a recognition pattern. The main control processor controls the recognition system, a supplementary light, and the camera. The supplementary light illuminates the recognition object and tracks the paint layer. The recognition system processes the data captured by the camera. A preset library in the recognition system contains multiple sets of feedback information, including image information and corresponding feedback results. The recognition pattern is compared with the image information in the preset library to confirm the feedback result corresponding to the recognition pattern. The advantages of this invention are that it prevents contact between the human body and methyl isobutylene ester, avoids errors caused by manual verification, eliminates the need to purchase a dedicated motion capture camera, and saves on operating costs.

Description

Technical Field

[0001] This invention belongs to the field of identification and marking systems, and in particular relates to a motion recognition system based on optical marking symbols. Background Technology

[0002] The earliest optical mark recognition (OMR) used paper tape and punched cards, with actual perforations on the punched cards, to read data by using light that passed through the paper.

[0003] Some OMR devices print markings on methyl isobutylene paper and interpret the data by measuring the intensity of light passing through the paper. However, methyl isobutylene itself is not conducive to human contact, and the transmittance is uncontrollable and easily affected by ambient light, thus limiting its application scenarios.

[0004] Modern OMR devices are mostly used for processing computer-generated images, such as barcodes, ticket counting, lottery tickets, and quizzes. However, these functions are applied to static image capture and processing, and cannot handle frame switching or motion processing, thus limiting their functionality to verification and inspection. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a motion recognition system based on optical marking symbols, which can prevent the human body from coming into contact with methyl isobutylene, avoid errors caused by manual verification, improve work efficiency, eliminate the need to purchase a dedicated motion capture camera, and save on usage costs.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a motion recognition system based on optical marking symbols, including a recognition system and a main control processor;

[0007] The recognition system includes a camera and a recognition object. The camera is used to capture the recognition object, and the recognition object is coated with paint, which forms a recognition pattern.

[0008] The main control processor is used to control the identification system, the fill light, and the camera;

[0009] The supplementary light is used to illuminate the object being identified and to track the coating.

[0010] The identification system is used to process the data captured by the camera. The preset library in the identification system contains multiple sets of feedback information, including image information and feedback results corresponding to the image information. The identification graphic is compared with the image information in the preset library to confirm the feedback result corresponding to the identification graphic.

[0011] Furthermore, the object to be identified is a punch card, which has positioning holes and pre-identification hole areas.

[0012] The positioning holes are used by the identification system to identify and locate the punched card and calculate the identification area;

[0013] The pre-identification hole area is located within the identification area. The pre-identification hole positions in the pre-identification hole area are coated with paint to form identification hole positions. The identification pattern is composed of one or more of the identification hole positions.

[0014] Furthermore, four positioning holes are provided, which are located at the four corners of the punch card, forming a rectangular recognition area.

[0015] Furthermore, the pre-identification hole area includes multiple pre-identification hole positions, which are distributed in 6 rows and 8 columns.

[0016] Furthermore, an auxiliary hole area is provided on the punch card to supplement and assist in the positioning of the positioning hole area.

[0017] Furthermore, the auxiliary hole area is provided in two sets, located on the upper and lower sides of the pre-identification hole area respectively.

[0018] Furthermore, the coating is a colored coating or a photosensitive coating.

[0019] Furthermore, the identification system's data processing flow for the camera includes receiving an MJPEG video stream, decoding and preprocessing each frame of data.

[0020] Furthermore, the preprocessing includes sequentially performing grayscale conversion, noise reduction, and binarization on the image.

[0021] Furthermore, the supplementary light is an infrared LED with a wavelength of 740nm.

[0022] An operation method for a motion recognition system based on optical marking symbols includes the following steps:

[0023] S1. Determine if it is a punch card:

[0024] The main control processor controls the fill light to emit a light source to illuminate the target. When the identification hole with paint is captured, the camera captures the identification hole in real time. The identification system processes the data captured by the camera to confirm that the target is a punch card.

[0025] When no identification hole with coating is detected, the identification system processes the data captured by the camera to confirm that the target is not a punch card.

[0026] S2. Capture the positioning hole position:

[0027] The identification system locks the coordinates of the positioning hole, determines whether the positioning hole matches the preset positioning coordinates in the preset library, calculates the identification area of ​​the positioning hole, identifies the identification hole within the identification area, and confirms the identification image;

[0028] S3. Confirm feedback results:

[0029] The recognition system includes multiple sets of feedback information in its preset library. The feedback information includes image information and feedback results corresponding to the image information. The recognition image is compared with the image information in the preset library to confirm the feedback result corresponding to the recognition image.

[0030] The advantages and positive effects of this invention are:

[0031] By adopting the above technical solution, the marking is avoided from being printed on paper made of methyl isobutyl ester, thus preventing human contact with methyl isobutyl ester and ensuring the safety of the use environment.

[0032] In the scoring process, errors caused by manual verification are avoided, thus improving work efficiency. The system enables rapid development of local identity verification functions, such as ticket verification and check-in, and allows for the design of corresponding identity passwords based on the different attributes of different individuals.

[0033] The supplementary lighting is suitable for different lighting environments, avoiding situations where verification is impossible due to insufficient light.

[0034] By controlling cameras through an identification system, there is no need to purchase dedicated motion capture cameras, saving on operating costs and enabling efficient development of related products. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the structure of the present invention;

[0036] Figure 2 This is a schematic diagram of the system of the present invention;

[0037] Figure 3 This is a data processing flowchart of the present invention;

[0038] Figure 4 This is a schematic diagram of the punch card of the present invention;

[0039] Figure 5 The present invention identifies the pre-image using the program.

[0040] Figure 6 This is a schematic diagram of the identification card surface of the present invention.

[0041] In the picture:

[0042] 1. Camera; 2. Main control processor;

[0043] 3. Punch-in card; 31. Auxiliary hole area; 311. Auxiliary hole position; 32. Positioning hole position; 33. Identification hole position; 34. Pre-identification hole area; 341. Pre-identification hole position;

[0044] 4. Identify the object;

[0045] 5. The program identifies the pre-image;

[0046] 6. Identify the card face. Detailed Implementation

[0047] The present invention will be further described below with reference to the embodiments and accompanying drawings.

[0048] like Figures 1 to 6 As shown, this embodiment provides a motion recognition system based on optical marking symbols, including a recognition system, a main control processor 2, and a supplementary light;

[0049] The recognition system includes a camera 1 and a recognition object 4. The camera 1 is used to capture the recognition object 4 in real time. The recognition object 4 is coated with paint, which is a colored or photosensitive paint, and the paint forms a unique recognition pattern.

[0050] The supplementary light is used to emit a light source to illuminate the identification object 4 and to track the paint on the identification object 4;

[0051] The identification system is used to process the data captured by camera 1. The identification system has a preset library with multiple sets of feedback information, including image information and feedback results corresponding to the image information. The identification pattern composed of paint is compared with the image information in the preset library to confirm the feedback result corresponding to the identification pattern.

[0052] The recognition system processes the captured data as follows: receiving the MJPEG video stream, decoding and preprocessing each frame of data. During the preprocessing of each frame of data, the image is sequentially converted to grayscale, denoised, and binarized. Denoising can be achieved using methods such as Gaussian filters, erosion, and dilation.

[0053] The main control processor 2 is used to control the camera 1, the fill light, and the recognition system.

[0054] The identification object 4 uses a punch card 3, and the supplementary light uses an infrared LED light with a wavelength of 740nm. The main control processor 2 is connected to the camera 1, and the main control processor 2 controls the LED light to illuminate the punch card 3.

[0055] The punch card 3 is provided with positioning holes 32, pre-identification hole area 34 and identification holes 33. The positioning holes 32 are used by the identification system to identify and locate the punch card 3, lock the coordinates of the positioning holes 32, determine whether they match the preset positioning coordinates in the preset library, and calculate the identification area. The pre-identification hole area 34 is located within the identification area and is the default hole position. The pre-identification holes 341 in the pre-identification hole area 34 are coated with colored or photosensitive paint to form identification holes 33, which are used to distinguish the pre-identification hole area 34. The identification pattern is composed of one or more identification holes 33. An auxiliary hole area 31 is also provided to supplement the pre-identification hole area 34 and to assist in the positioning of the positioning holes 32.

[0056] When the object 4 (i.e., the punch card 3) captured by the camera 1 is in motion, if the punch card 3 is always within the capture range of the camera 1, the identification system locks the coordinates of the positioning hole 32 frame by frame. Based on the changes in the image coordinates of the previous and next frames, the system judges the difference between the previous and next frames, obtains the distance and direction of the object 4's movement, and further calculates the moving speed and trajectory of the object 4.

[0057] Example 1:

[0058] The punch card 3 is provided with positioning holes 32 and pre-identification hole area 34. There are four positioning holes 32, which are spaced at a fixed distance and are located at the four corners of the punch card 3. The four positioning holes 32 form a rectangular identification area.

[0059] The pre-identification hole area 34 is located within the rectangular identification area. The pre-identification hole area 34 includes multiple pre-identification hole positions 341, which are arranged in 6 rows and 8 columns, for a total of 48.

[0060] The pre-identification hole positions 341 on the pre-identification hole area 34 are coated with colored or photosensitive paint to form identification hole positions 33. The identification hole positions 33 are recorded in coordinate form. Multiple identification hole positions 33 or a single identification hole position 33 form an identification pattern for comparison with a preset library in the identification system.

[0061] Example 2:

[0062] The punch card 3 is provided with positioning holes 32, pre-identification hole area 34 and auxiliary hole area 31.

[0063] There are four positioning holes 32, which are spaced at a fixed distance and are located at the four corners of the punch card 3, forming a rectangular recognition area.

[0064] The pre-identification hole area 34 is located within a rectangular identification area. The pre-identification hole area 34 includes multiple pre-identification hole positions 341, which are arranged in 6 rows and 8 columns, for a total of 48 holes.

[0065] The auxiliary hole area 31 is provided in two groups, located on the upper and lower sides of the pre-identification hole area 34 respectively. The auxiliary hole positions 311 in one group are distributed in 2 rows and 6 columns. There are a total of 12 auxiliary hole positions 311 in one group, and a total of 24 auxiliary hole positions 311 in both groups.

[0066] The pre-identification hole positions 341 on the pre-identification hole area 34 are coated with colored or photosensitive paint to form identification hole positions 33. The identification hole positions 33 are recorded in coordinate form. Multiple identification hole positions 33 or a single identification hole position 33 form an identification pattern for comparison with a preset library in the identification system.

[0067] The working process of this example:

[0068] The main control processor 2 controls the fill light to emit a light source to illuminate the target. The fill light uses an LED light with a wavelength of 740nm. When the camera 1 captures the identification hole 33 coated with colored or photosensitive paint, the camera 1 captures the identification hole 33 in real time and tracks the movement of the image by continuously updating the still frame (FPS30). The identification system confirms that the target captured by the camera 1 is the punch card 3 through calculation and processing.

[0069] If the identification hole 33 coated with colored or photosensitive paint is not captured, the identification system will calculate and confirm that the object captured by the camera 1 is not the punch card 3.

[0070] The identification process of the recognition system for capturing objects is as follows: receiving MJPEG video stream, decoding and preprocessing each frame of data. The preprocessing involves performing grayscale processing, noise reduction processing and binarization processing on the image in sequence. Noise reduction processing can use methods such as Gaussian filter, erosion, and dilation.

[0071] After capturing the punch card 3, the positioning hole 32 is captured first, the coordinates of the positioning hole 32 are locked, and it is determined whether the positioning hole 32 conforms to the preset positioning coordinates in the preset library. According to the calculation, the four points of the four positioning holes 32 are connected to form a rectangular recognition area. Within the recognition area, it is determined whether the recognition pattern formed by the recognition holes 33 corresponds to the image information in the preset library. If it corresponds to a certain image information in the preset library, the feedback result corresponding to the image information is obtained.

[0072] In this embodiment, two identification holes 33 are provided in the pre-identification hole area 34. The first identification hole is located in the fourth row and seventh column, and its coordinates are expressed as fourth row (0,0,0,1,0,0), seventh column (0,0,0,0,0,0,1,0). The second identification hole is located in the sixth row and sixth column, and its coordinates are expressed as sixth row (0,0,0,0,0,1), sixth column (0,0,0,0,0,1,0,0). The identification system searches and matches the identification image formed by combining the two identification holes with the image information in the preset library. In the preset library, there is graphic information composed of the first identification hole in the fourth row and seventh column and the second identification hole in the sixth row and sixth column. The feedback result of the graphic information is 606. Therefore, through the punch card 3 in this embodiment, through the calculation and fault tolerance processing of the identification system, the feedback result is also 606.

[0073] If the positioning hole 32 is always within the shooting range of the camera 1, the positioning hole 32 can be locked frame by frame by the recognition system to ensure real-time identification and positioning of the punch card 3. The difference between the previous and previous frames can be judged based on the coordinate changes of the punch card 3 identified in the previous and previous frames, so as to further calculate the distance and direction and obtain the moving speed and trajectory of the punch card 3.

[0074] The beneficial effects of this invention are:

[0075] This invention solves the problem of errors caused by manual verification in scoring work and improves work efficiency.

[0076] By coating the pre-identification hole positions 341 on the pre-identification hole area 34 with colored or photosensitive paint to form identification hole positions 33, the marking is avoided from being printed on paper made of methyl isobutyl acrylate. By arranging and combining different positions of the identification hole positions 33, different feedback results can be obtained, allowing more sets of feedback information to be stored in the preset library. This assists in setting the hole area 31, and serves to expand the total count when the pre-identification hole positions 341 are insufficient, and can be used when the identification environment is poor and identification positioning needs to be strengthened, increasing the number of feedback results and making it more widely applicable.

[0077] By using LED lights, the problem of poor recognition performance under different lighting conditions is solved. Furthermore, by using 740nm wavelength infrared light, the influence of natural light on the recognition light is greatly avoided, and verification can be completed even in insufficient lighting conditions.

[0078] The identification system solves the cost problem of developing recognition and tracking functions in artificial intelligence modules, eliminating the need to purchase dedicated motion capture camera equipment and enabling efficient development of related products.

[0079] It can quickly identify local identity verification functions and can be applied to ticket verification, access check-in and other occasions. It can also form unique identity passwords based on the different attributes of different people.

[0080] The foregoing has described one or more embodiments of the present invention in detail, but the description is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A motion recognition system based on optical marking symbols, characterized in that: Including the identification system and the main control processor (2); The recognition system includes a camera (1) and a recognition object (4). The camera (1) is used to capture the recognition object (4). The recognition object (4) is coated with paint, and the paint forms a recognition pattern. The main control processor (2) is used to control the identification system, the fill light and the camera (1); The supplementary light is used to illuminate the identification object (4) and track the coating; The identification system is used to process the data captured by the camera (1). The preset library in the identification system is set with multiple sets of feedback information, including image information and feedback results corresponding to the image information. The identification graphic is compared with the image information in the preset library to confirm the feedback result corresponding to the identification graphic. The identification object (4) is a punch card (3), and the punch card (3) is provided with positioning holes (32) and pre-identification hole area (34). The positioning hole (32) is used by the identification system to identify and locate the punch card (3) and calculate the identification area; The pre-identification hole area (34) is located within the identification area. The pre-identification hole positions (341) on the pre-identification hole area (34) are coated with paint to form identification hole positions (33). The identification pattern is composed of one or more of the identification hole positions (33). The identification system locks the coordinates of the positioning hole (32) frame by frame, judges the difference between the previous and previous frames based on the changes in the image coordinates identified in the previous and previous frames, obtains the distance and direction of the movement of the identified object (4), and further calculates the movement speed and trajectory of the identified object (4). An auxiliary hole area (31) is provided on the punch card (3) to supplement the pre-identification hole area (34) and assist in the positioning of the positioning hole (32).

2. The motion recognition system based on optical marking symbols according to claim 1, characterized in that: The coating is a colored coating or a photosensitive coating.

3. The motion recognition system based on optical marking symbols according to claim 1, characterized in that: The identification system's data processing flow for the camera (1) includes receiving MJPEG video streams, decoding and preprocessing each frame of data.

4. The motion recognition system based on optical marking symbols according to claim 3, characterized in that: The preprocessing includes sequentially performing grayscale conversion, noise reduction, and binarization on the image.

5. A motion recognition system based on optical marking symbols according to claim 1, characterized in that: The supplementary light uses an infrared LED with a wavelength of 740nm.

6. An operating method for the motion recognition system as described in any one of claims 1 to 5, comprising the following steps: S1. Determine if it is a punch card (3): The main control processor (2) controls the fill light to emit light source to illuminate the captured object. When the identification hole (33) coated with paint is captured, the camera (1) captures the identification hole (33) in real time. The identification system processes the data captured by the camera (1) to confirm that the captured object is a punch card (3). When the identification hole (33) coated with paint is not captured, the identification system processes the data captured by the camera (1) to confirm that the captured object is not a punch card (3). S2. Capture the positioning hole (32): The identification system locks the coordinates of the positioning hole (32), determines whether the positioning hole (32) conforms to the preset positioning coordinates in the preset library, calculates the identification area of ​​the positioning hole (32), identifies the identification hole (33) in the identification area, and confirms the identification graphic; S3. Confirm feedback results: The recognition system includes multiple sets of feedback information in its preset library. The feedback information includes image information and feedback results corresponding to the image information. The recognition image is compared with the image information in the preset library to confirm the feedback result corresponding to the recognition image.

Images