A method and system for intelligently identifying and locating irregular targets

Abstract

The present invention discloses a method and system for intelligently identifying and locating irregular targets, which are applied to the field of laser marking technology, construct a three-dimensional database, and the three-dimensional database stores three-dimensional data of the target; obtain a three-dimensional image of the target to be identified; extract features from the three-dimensional image to obtain a first positioning feature, and call the positioning feature of the target point; input the first positioning feature into a trained positioning detection model, adjust the position information of the positioning point according to the reflection feature and color feature of the target to be identified, and compare it with the positioning feature; determine the position of the positioning point. The present invention discloses a method and system for intelligently identifying and locating irregular targets, which corrects the acquired reflection feature through the light reflection detection result, clusters similar colors with the color detection result output by the color detection model, determines the laser marking range, and the key point detection result output by the key point detection model jointly locates the key point.

Description

Technical Field

[0001] The present invention relates to the field of laser marking technology, and more particularly to a method and system for intelligently identifying and locating irregular targets. Background Art

[0002] Laser marking technology can not only ensure that the integrity of metal workpieces is not damaged, but also etch information patterns that can be permanently retained. It is very suitable for manufacturing industries with high precision requirements.

[0003] With the vigorous development of the Internet of Things industry and manufacturing industry at home and abroad, QR code technology has gradually been used in the metal parts identification in the manufacturing field due to its powerful error correction ability. In order to improve the management efficiency of enterprises, laser QR code has been widely used as one of the effective technologies to solve the problem of metal parts identification, that is, using laser marking technology to directly mark the metal parts with the corresponding information QR code. However, the current use of image processing technology to identify laser QR codes on the surface of metal parts has problems such as positioning difficulties. The main reasons are: (1) The outer surface of the metal workpiece is not completely flat, which causes the QR code image to have a curvature during laser direct marking; (2) The materials of the metal workpiece itself vary. Some metal parts have reflective surfaces, some are dark in color, and some are easily oxidized, which increases the difficulty of accurately locating and identifying the QR code; (3) The materials of metal workpieces are different. During the laser direct marking process, QR codes with black as the foreground color cannot be directly marked, which is exactly the opposite of the standard QR code; (4) When marking the QR code on the metal workpiece, the internal factors and external factors of the laser coder cause the QR code presented on the tool surface to be irregular; (5) The recognition deviation of the target leads to positioning errors.

[0004] Therefore, how to provide a method and system for accurately identifying and locating a target is an urgent problem to be solved by those skilled in the art. Summary of the invention

[0005] In view of this, the present invention provides an irregular target intelligent recognition and positioning method and system. For the recognition of irregular targets, the accuracy is greatly improved by calling historical data for comparison. For the determination of the marking position, the reflection detection model, color detection model and the key point detection model are used to greatly improve the positioning accuracy.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] An irregular target intelligent recognition and positioning method, the specific steps are as follows:

[0008] Building a three-dimensional database, wherein the three-dimensional database stores target three-dimensional data;

[0009] Acquire a three-dimensional image of a target to be identified;

[0010] Performing feature extraction on the three-dimensional image to obtain a first positioning feature, and calling the positioning feature of the target point;

[0011] Inputting the first positioning feature into the trained positioning detection model, adjusting the position information of the positioning point according to the reflection feature and color feature of the target to be identified, and comparing it with the positioning feature;

[0012] Determine the location of the anchor point.

[0013] Optionally, in the above-mentioned method for intelligently identifying and locating irregular targets, the specific steps of obtaining a three-dimensional image of the target to be identified include:

[0014] Obtain at least 3 2D images from different angles;

[0015] According to the correspondence between the three-dimensional data and the two-dimensional coordinates of the acquired two-dimensional image and the shooting angle information of each picture, the three-dimensional coordinate values ​​of all points corresponding to the pixel points of the key points on each two-dimensional image are obtained;

[0016] According to the predetermined rules, a specific three-dimensional coordinate value is selected as the coordinate of the three-dimensional image space.

[0017] The coordinates generate a binary three-dimensional image to determine the target to be identified.

[0018] Optionally, in the above-mentioned method for intelligently identifying and locating irregular targets, the steps for constructing the positioning detection model are as follows:

[0019] Acquire a first sample image set; the first sample image set includes a plurality of first sample images, each of the first sample images is associated with a key point expected result;

[0020] The following operations are performed for each of the first sample images to obtain the key point loss value of the first sample image:

[0021] The first sample image is processed using a reflection detection model, a color detection model and the key point detection model respectively to obtain a light reflection detection result output by the reflection detection model, a color detection result output by the color detection model and a key point detection result output by the key point detection model;

[0022] Determining key point loss information between the key point detection result and the key point expected result;

[0023] The key point loss information is adjusted by using the light reflection detection result and the color detection result to obtain the key point loss value of the first sample image;

[0024] Using the key point loss values ​​of each of the first sample images to determine a first total loss value;

[0025] Optimizing parameters of the key point detection model according to the first total loss value.

[0026] Optionally, in the above-mentioned method for intelligently identifying and locating irregular targets, the steps for constructing the reflection detection model are as follows:

[0027] Calculate the viewpoint diffuse reflection data and the viewpoint highlight reflection data corresponding to the key point in the current viewpoint direction according to the current viewpoint direction information;

[0028] A diffuse reflection weight corresponding to the viewpoint diffuse reflection data and a highlight reflection weight corresponding to the viewpoint highlight reflection data are obtained, and the fill light reflection data is determined by weighted summing of the viewpoint diffuse reflection data and the viewpoint highlight reflection data.

[0029] Optionally, in the above-mentioned method for intelligently identifying and locating irregular targets, the steps for constructing the color detection model are as follows:

[0030] Determine the attribute average value of the key point, find the eight-neighborhood pixels of the key point, and judge the similarity between the attributes of the eight-neighborhood pixels and the attribute average value, and judge whether the similarity is greater than a preset threshold value, when it is greater than the preset threshold value, determine the eight-neighborhood pixels as the first key point, and determine the area where the first key point is located as the first key area, when it is less than the preset threshold value, merge the eight-neighborhood pixels into the key area, and replace the original key point in the key area with the eight-neighborhood pixels to become the key point of the key area;

[0031] Traverse all key points to determine the laser marking range.

[0032] An irregular target intelligent recognition and positioning system, comprising:

[0033] A three-dimensional database storing target three-dimensional data;

[0034] An acquisition module, which acquires a three-dimensional image of a target to be identified;

[0035] A feature extraction module performs feature extraction on the three-dimensional image to obtain a first positioning feature, and calls the positioning feature of the target point;

[0036] The positioning module inputs the first positioning feature into the trained positioning detection model, adjusts the position information of the positioning point according to the reflection feature and color feature of the target to be identified, and compares it with the positioning feature; and determines the position of the positioning point.

[0037] Optionally, in the above-mentioned irregular target intelligent recognition and positioning system, the acquisition module includes:

[0038] A two-dimensional image unit, which obtains at least three two-dimensional images at different angles;

[0039] A coordinate conversion unit, which obtains the three-dimensional coordinate values ​​of all points corresponding to the pixel points of the key points on each two-dimensional image according to the corresponding relationship between the three-dimensional data and the two-dimensional coordinates of the acquired two-dimensional image and the shooting angle information of each image;

[0040] The recognition unit selects specific three-dimensional coordinate values ​​as coordinates of the three-dimensional image space according to a predetermined rule, generates a binary three-dimensional image from the coordinates, and determines the target to be recognized.

[0041] Optionally, in the above-mentioned irregular target intelligent recognition and positioning system, the positioning module includes:

[0042] A sample set unit is configured to obtain a first sample image set; the first sample image set includes a plurality of first sample images, each of which is associated with a key point expected result;

[0043] The loss value unit performs the following operations on each of the first sample images to obtain a key point loss value of the first sample image:

[0044] a detection unit, which processes the first sample image using a reflection detection model, a color detection model, and the key point detection model, respectively, to obtain a light reflection detection result output by the reflection detection model, a color detection result output by the color detection model, and a key point detection result output by the key point detection model;

[0045] an optimization unit, determining key point loss information between the key point detection result and the key point expected result; adjusting the key point loss information by using the light reflection detection result and the color detection result to obtain the key point loss value of the first sample image; and determining a first total loss value by using the key point loss values ​​of each of the first sample images;

[0046] Optimizing parameters of the key point detection model according to the first total loss value.

[0047] Optionally, in the above-mentioned irregular target intelligent recognition and positioning system, the reflection detection model includes:

[0048] Calculate the viewpoint diffuse reflection data and the viewpoint highlight reflection data corresponding to the key point in the current viewpoint direction according to the current viewpoint direction information;

[0049] A diffuse reflection weight corresponding to the viewpoint diffuse reflection data and a highlight reflection weight corresponding to the viewpoint highlight reflection data are obtained, and the fill light reflection data is determined by weighted summing of the viewpoint diffuse reflection data and the viewpoint highlight reflection data.

[0050] Optionally, in the above-mentioned irregular target intelligent recognition and positioning system, the color detection model includes:

[0051] Determine the attribute average value of the key point, find the eight-neighborhood pixels of the key point, and judge the similarity between the attributes of the eight-neighborhood pixels and the attribute average value, and judge whether the similarity is greater than a preset threshold value, when it is greater than the preset threshold value, determine the eight-neighborhood pixels as the first key point, and determine the area where the first key point is located as the first key area, when it is less than the preset threshold value, merge the eight-neighborhood pixels into the key area, and replace the original key point in the key area with the eight-neighborhood pixels to become the key point of the key area;

[0052] Traverse all key points to determine the laser marking range.

[0053] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method and system for intelligent identification and positioning of irregular targets. For the identification of irregular targets, the accuracy is greatly improved by calling historical data for comparison. The positioning accuracy is greatly improved by using the reflection detection model, the color detection model and the key point detection model to determine the marking position. The acquired reflection features are corrected by the light reflection detection results, the color detection results output by the color detection model cluster similar colors, and the laser marking range and the key point detection results output by the key point detection model are used to locate the key points together. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0055] Figure 1 is a flow chart of the method of the present invention;

[0056] Figure 2 It is a structural block diagram of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The embodiment of the present invention discloses a method for intelligently identifying and locating irregular targets. Figure 1 As shown, the specific steps are as follows:

[0059] Building a three-dimensional database, the three-dimensional database stores target three-dimensional data;

[0060] Acquire a three-dimensional image of a target to be identified;

[0061] Extract features from the three-dimensional image to obtain a first positioning feature, and call the positioning feature of the target point;

[0062] Inputting the first positioning feature into the trained positioning detection model, adjusting the position information of the positioning point according to the reflection feature and color feature of the target to be identified, and comparing it with the positioning feature;

[0063] Determine the location of the anchor point.

[0064] In order to further optimize the above technical solution, the specific steps of obtaining a three-dimensional image of the target to be identified include:

[0065] Obtain at least 3 2D images from different angles;

[0066] According to the correspondence between the three-dimensional data and the two-dimensional coordinates of the acquired two-dimensional image and the shooting angle information of each picture, the three-dimensional coordinate values ​​of all points corresponding to the pixel points of the key points on each two-dimensional image are obtained;

[0067] According to predetermined rules, specific three-dimensional coordinate values ​​are selected as coordinates of the three-dimensional image space, and a binary three-dimensional image is generated from the coordinates to determine the target to be identified.

[0068] Furthermore, three-dimensional scanning can be performed to obtain three-dimensional images;

[0069] Furthermore, the two-dimensional image is reconstructed into a three-dimensional image, and the three-dimensional image is reconstructed according to the shooting angle information; the two-dimensional image is easy to obtain, the equipment requirements are greatly reduced, and costs can be saved.

[0070] In order to further optimize the above technical solution, the steps for building the positioning detection model are as follows:

[0071] Acquire a first sample image set; the first sample image set includes a plurality of first sample images, each of which is associated with a key point expected result;

[0072] The following operations are performed for each first sample image to obtain the key point loss value of the first sample image:

[0073] The first sample image is processed using a reflection detection model, a color detection model, and a key point detection model respectively to obtain a light reflection detection result output by the reflection detection model, a color detection result output by the color detection model, and a key point detection result output by the key point detection model;

[0074] Determine the key point loss information between the key point detection result and the key point expected result;

[0075] The key point loss information is adjusted using the light reflection detection result and the color detection result to obtain the key point loss value of the first sample image;

[0076] Using the key point loss values ​​of each first sample image to determine a first total loss value;

[0077] Optimize the parameters of the keypoint detection model according to the first total loss value.

[0078] Furthermore, the acquired reflection features are corrected through the light reflection detection results, the color detection results output by the color detection model cluster similar colors, determine the laser marking range, and the key point detection results output by the key point detection model jointly locate the key points.

[0079] In order to further optimize the above technical solution, the steps for building the reflection detection model are as follows:

[0080] According to the current viewpoint direction information, the viewpoint diffuse reflection data and the viewpoint highlight reflection data corresponding to the key point in the current viewpoint direction are calculated;

[0081] The diffuse reflection weight corresponding to the viewpoint diffuse reflection data and the highlight reflection weight corresponding to the viewpoint highlight reflection data are obtained, and the fill light reflection data is determined by weighted summing of the viewpoint diffuse reflection data and the viewpoint highlight reflection data.

[0082] Due to the irregularity of the target, different colors and curvatures will reflect light differently. The light reflection is compensated for to ensure accurate positioning.

[0083] In order to further optimize the above technical solution, the steps for building the color detection model are as follows:

[0084] Determine the attribute average value of the key point, find the eight-neighborhood pixels of the key point, and judge the similarity between the attributes of the eight-neighborhood pixels and the attribute average value, and judge whether the similarity is greater than a preset threshold. When it is greater than the preset threshold, determine the eight-neighborhood pixels as the first key point, and determine the area where the first key point is located as the first key area. When it is less than the preset threshold, merge the eight-neighborhood pixels into the key area, and replace the original key point in the key area with the eight-neighborhood pixels to become the key point of the key area.

[0085] Traverse all key points to determine the laser marking range.

[0086] Another embodiment of the present invention discloses an irregular target intelligent recognition and positioning system, such as Figure 2 As shown, including,

[0087] A three-dimensional database stores target three-dimensional data;

[0088] An acquisition module, which acquires a three-dimensional image of a target to be identified;

[0089] A feature extraction module extracts features from the three-dimensional image to obtain a first positioning feature, and calls the positioning feature of the target point;

[0090] The positioning module inputs the first positioning feature into the trained positioning detection model, adjusts the position information of the positioning point according to the reflection feature and color feature of the target to be identified, and compares it with the positioning feature; and determines the position of the positioning point.

[0091] In order to further optimize the above technical solution, the acquisition module includes:

[0092] A two-dimensional image unit, which obtains at least three two-dimensional images at different angles;

[0093] A coordinate conversion unit, which obtains the three-dimensional coordinate values ​​of all points corresponding to the pixel points of the key points on each two-dimensional image according to the corresponding relationship between the three-dimensional data and the two-dimensional coordinates of the acquired two-dimensional image and the shooting angle information of each image;

[0094] The recognition unit selects specific three-dimensional coordinate values ​​as coordinates of the three-dimensional image space according to predetermined rules, generates a binary three-dimensional image from the coordinates, and determines the target to be recognized.

[0095] In order to further optimize the above technical solution, the positioning module includes:

[0096] The sample set unit acquires a first sample image set; the first sample image set includes a plurality of first sample images, each of which is associated with a key point expected result;

[0097] The loss value unit performs the following operations on each first sample image to obtain the key point loss value of the first sample image:

[0098] a detection unit, which processes the first sample image using a reflection detection model, a color detection model, and a key point detection model, respectively, to obtain a light reflection detection result output by the reflection detection model, a color detection result output by the color detection model, and a key point detection result output by the key point detection model;

[0099] The optimization unit determines key point loss information between the key point detection result and the key point expected result; uses the light reflection detection result and the color detection result to adjust the key point loss information to obtain the key point loss value of the first sample image; uses the key point loss values ​​of each first sample image to determine the first total loss value;

[0100] Optimize the parameters of the keypoint detection model according to the first total loss value.

[0101] In order to further optimize the above technical solution, the reflection detection model includes:

[0102] According to the current viewpoint direction information, the viewpoint diffuse reflection data and the viewpoint highlight reflection data corresponding to the key point in the current viewpoint direction are calculated;

[0103] The diffuse reflection weight corresponding to the viewpoint diffuse reflection data and the highlight reflection weight corresponding to the viewpoint highlight reflection data are obtained, and the fill light reflection data is determined by weighted summing of the viewpoint diffuse reflection data and the viewpoint highlight reflection data.

[0104] In order to further optimize the above technical solution, the color detection model includes:

[0105] Determine the attribute average value of the key point, find the eight-neighborhood pixels of the key point, and judge the similarity between the attributes of the eight-neighborhood pixels and the attribute average value, and judge whether the similarity is greater than a preset threshold. When it is greater than the preset threshold, determine the eight-neighborhood pixels as the first key point, and determine the area where the first key point is located as the first key area. When it is less than the preset threshold, merge the eight-neighborhood pixels into the key area, and replace the original key point in the key area with the eight-neighborhood pixels to become the key point of the key area.

[0106] Traverse all key points to determine the laser marking range.

[0107] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0108] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for intelligent recognition and positioning of irregular targets, characterized in that: The specific steps are as follows: Building a three-dimensional database, wherein the three-dimensional database stores target three-dimensional data; Acquire a three-dimensional image of a target to be identified; Performing feature extraction on the three-dimensional image to obtain a first positioning feature, and calling the positioning feature of the target point; Inputting the first positioning feature into the trained positioning detection model, adjusting the position information of the positioning point according to the reflection feature and color feature of the target to be identified, and comparing it with the positioning feature; Determine the location of the anchor point; The steps for constructing the positioning detection model are as follows: Acquire a first sample image set; the first sample image set includes a plurality of first sample images, each of the first sample images is associated with a key point expected result; The following operations are performed for each of the first sample images to obtain the key point loss value of the first sample image: The first sample image is processed using a reflection detection model, a color detection model and the key point detection model respectively to obtain a light reflection detection result output by the reflection detection model, a color detection result output by the color detection model and a key point detection result output by the key point detection model; Determining key point loss information between the key point detection result and the key point expected result; The key point loss information is adjusted by using the light reflection detection result and the color detection result to obtain the key point loss value of the first sample image; Using the key point loss values ​​of each of the first sample images to determine a first total loss value; Optimizing parameters of the key point detection model according to the first total loss value.

2. The method for intelligently identifying and locating irregular targets according to claim 1, characterized in that: The specific steps of obtaining a three-dimensional image of the target to be identified include: Obtain at least 3 2D images from different angles; According to the correspondence between the three-dimensional data and the two-dimensional coordinates of the acquired two-dimensional image and the shooting angle information of each picture, the three-dimensional coordinate values ​​of all points corresponding to the pixel points of the key points on each two-dimensional image are obtained; According to a predetermined rule, a specific three-dimensional coordinate value is selected as the coordinate of the three-dimensional image space, a binary three-dimensional image is generated from the coordinate, and the target to be identified is determined.

3. The method for intelligently identifying and locating irregular targets according to claim 1, characterized in that: The steps for constructing the reflection detection model are as follows: Calculate the viewpoint diffuse reflection data and the viewpoint highlight reflection data corresponding to the key point in the current viewpoint direction according to the current viewpoint direction information; A diffuse reflection weight corresponding to the viewpoint diffuse reflection data and a highlight reflection weight corresponding to the viewpoint highlight reflection data are obtained, and fill light reflection data are determined by weighted summing of the viewpoint diffuse reflection data and the viewpoint highlight reflection data.

4. The method for intelligently identifying and locating irregular targets according to claim 1, characterized in that: The steps for constructing the color detection model are as follows: Determine the attribute average value of the key point, find the eight-neighborhood pixels of the key point, and judge the similarity between the attributes of the eight-neighborhood pixels and the attribute average value, and judge whether the similarity is greater than a preset threshold value, when it is greater than the preset threshold value, determine the eight-neighborhood pixels as the first key point, and determine the area where the first key point is located as the first key area, when it is less than the preset threshold value, merge the eight-neighborhood pixels into the key area, and replace the original key point in the key area with the eight-neighborhood pixels to become the key point of the key area; Traverse all key points to determine the laser marking range.

5. An irregular target intelligent recognition and positioning system, characterized in that: include: A three-dimensional database storing target three-dimensional data; An acquisition module, which acquires a three-dimensional image of a target to be identified; A feature extraction module performs feature extraction on the three-dimensional image to obtain a first positioning feature, and calls the positioning feature of the target point; A positioning module, inputting the first positioning feature into a trained positioning detection model, adjusting the position information of the positioning point according to the reflection feature and color feature of the target to be identified, and comparing it with the positioning feature; determining the position of the positioning point; The steps for constructing the positioning detection model are as follows: Acquire a first sample image set; the first sample image set includes a plurality of first sample images, each of the first sample images is associated with a key point expected result; The following operations are performed for each of the first sample images to obtain the key point loss value of the first sample image: The first sample image is processed using a reflection detection model, a color detection model and the key point detection model respectively to obtain a light reflection detection result output by the reflection detection model, a color detection result output by the color detection model and a key point detection result output by the key point detection model; Determining key point loss information between the key point detection result and the key point expected result; The key point loss information is adjusted by using the light reflection detection result and the color detection result to obtain the key point loss value of the first sample image; Using the key point loss values ​​of each of the first sample images to determine a first total loss value; Optimizing parameters of the key point detection model according to the first total loss value.

6. The irregular target intelligent recognition and positioning system according to claim 5, characterized in that: The acquisition modules include: A two-dimensional image unit, which obtains at least three two-dimensional images at different angles; A coordinate conversion unit, which obtains the three-dimensional coordinate values ​​of all points corresponding to the pixel points of the key points on each two-dimensional image according to the corresponding relationship between the three-dimensional data and the two-dimensional coordinates of the acquired two-dimensional image and the shooting angle information of each image; The recognition unit selects specific three-dimensional coordinate values ​​as coordinates of the three-dimensional image space according to a predetermined rule, generates a binary three-dimensional image from the coordinates, and determines the target to be recognized.

7. The irregular target intelligent recognition and positioning system according to claim 5, characterized in that: The positioning module comprises: A sample set unit is configured to obtain a first sample image set; the first sample image set includes a plurality of first sample images, each of which is associated with a key point expected result; The loss value unit performs the following operations on each of the first sample images to obtain a key point loss value of the first sample image: a detection unit, which processes the first sample image using a reflection detection model, a color detection model, and the key point detection model, respectively, to obtain a light reflection detection result output by the reflection detection model, a color detection result output by the color detection model, and a key point detection result output by the key point detection model; an optimization unit, determining key point loss information between the key point detection result and the key point expected result; adjusting the key point loss information by using the light reflection detection result and the color detection result to obtain the key point loss value of the first sample image; and determining a first total loss value by using the key point loss values ​​of each of the first sample images; Optimizing parameters of the key point detection model according to the first total loss value.

8. The irregular target intelligent recognition and positioning system according to claim 7, characterized in that: The reflection detection model includes: Calculate the viewpoint diffuse reflection data and the viewpoint highlight reflection data corresponding to the key point in the current viewpoint direction according to the current viewpoint direction information; A diffuse reflection weight corresponding to the viewpoint diffuse reflection data and a highlight reflection weight corresponding to the viewpoint highlight reflection data are obtained, and fill light reflection data are determined by weighted summing of the viewpoint diffuse reflection data and the viewpoint highlight reflection data.

9. The irregular target intelligent recognition and positioning system according to claim 7, characterized in that: The color detection model includes: Determine the attribute average value of the key point, find the eight-neighborhood pixels of the key point, and judge the similarity between the attributes of the eight-neighborhood pixels and the attribute average value, and judge whether the similarity is greater than a preset threshold value, when it is greater than the preset threshold value, determine the eight-neighborhood pixels as the first key point, and determine the area where the first key point is located as the first key area, when it is less than the preset threshold value, merge the eight-neighborhood pixels into the key area, and replace the original key point in the key area with the eight-neighborhood pixels to become the key point of the key area; Traverse all key points to determine the laser marking range.

Images