Machine vision method

Abstract

A machine vision method belongs to the technical field of machine vision, and comprises the following steps: initially obtaining an orientation value of a camera, forming a decoding coefficient I according to the orientation value of the camera, and forming a coding coefficient I according to the decoding coefficient I; then obtaining a bit rate critical quantity when the camera transmits product image information to the image controller, forming a decoding coefficient 2 according to the bit rate critical quantity, and forming a coding coefficient 2 according to the decoding coefficient 2; then, a third decoding coefficient of the camera is formed according to the first decoding coefficient and the second decoding coefficient, and a third coding coefficient is formed according to the third decoding coefficient; and finally, when the image controller receives the product image information transmitted by the camera, the first coding coefficient, the second coding coefficient and the third coding coefficient are started successively, and whether the received product image information is processed or not is determined successively according to the first decoding coefficient, the second decoding coefficient and the third decoding coefficient. And the reliability of the image controller during product image information processing can be greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine vision, and in particular relates to a machine vision method. Background Art

[0002] Machine vision is a rapidly developing branch of artificial intelligence. Simply put, it uses machines to replace the human eye for measurement and identification. A machine vision system uses a machine vision product (i.e., an image capture device, available in CMOS and CCD) to convert captured objects into image signals. This signal is then transmitted to a dedicated image processing system, which obtains the object's morphological information and converts it into a digital signal based on pixel distribution, brightness, color, and other information. The imaging system then performs various operations on these signals to extract the target's features, and uses this information to control the operation of on-site equipment.

[0003] During machine vision applications, the existing technical solution with patent application number "CN201020695088.0" and patent name "A robotic arm structure with machine vision navigation function" is often used to achieve machine vision applications. The patent solution includes a camera for capturing product images and an image controller for processing product images. The camera and the image controller are connected, and the camera transmits the captured product images to the image controller for processing.

[0004] However, there are quite a few cameras, and the image controller must generally determine the rationality of the source of the product image information based on the reliability of the product image information when receiving and processing the product image information. When determining the rationality of the source of the product image information, the current method is to process the product image information first, and then extract the source information of the product image information. If the source information of the product image information is the same as that stored in advance, it is confirmed that the source of the product image information is reasonable. However, if the product image information processing is performed first, if the product image information received by the image controller has its own detection process, in case the product image information is processed first, the detection process will often be forced to be processed or the product image information will be abnormally attacked, so that the reliability of the image controller is not guaranteed, and the product image information stored in the image controller will often be stolen. Summary of the Invention

[0005] In order to solve the defects existing in the prior art, the present invention proposes a machine vision method, which initially obtains the orientation value of the camera, and forms a decoding coefficient one based on the orientation value of the camera, and forms an encoding coefficient one based on the decoding coefficient one; then obtains the bit rate critical value when the camera transmits product image information to the image controller, and forms a decoding coefficient two based on the bit rate critical value, and forms an encoding coefficient two based on the decoding coefficient two; then, forms a decoding coefficient three of the camera based on the decoding coefficient one and the decoding coefficient two, and forms an encoding coefficient three based on the decoding coefficient three; finally, when the image controller receives the product image information transmitted by the camera, it successively activates the encoding coefficient one, the encoding coefficient two and the encoding coefficient three, and successively activates the encoding coefficient one, the encoding coefficient two and the encoding coefficient three. Based on decoding coefficient one, decoding coefficient two and decoding coefficient three, it is determined whether to process the received product image information. For the use of encoding and decoding coefficients, decoding coefficient one is formed based on the orientation value of the camera, and decoding coefficient two is formed based on the bit rate threshold. In addition, decoding coefficient three must be formed together with decoding coefficient one and decoding coefficient two. Therefore, when the image controller performs the rationality of the image information source later, it must ensure that the corresponding encoding coefficient one, encoding coefficient two and encoding coefficient three are all reasonable before enabling decoding coefficient one, decoding coefficient two and decoding coefficient three to process the received product image information. Therefore, the reliability of the image controller in processing product image information can be greatly improved.

[0006] The present invention utilizes the following technical solutions.

[0007] A machine vision method, comprising:

[0008] The camera transmits the captured product image to the image controller for processing, which includes:

[0009] Step 1: Obtain the camera's position value, generate a decoding coefficient 1 based on the camera's position value, and generate an encoding coefficient 1 based on the decoding coefficient 1;

[0010] Step 2: Obtaining a bit rate threshold when the camera transmits the product image to the image controller, generating a decoding coefficient 2 according to the bit rate threshold, and generating an encoding coefficient 2 according to the decoding coefficient 2;

[0011] Step 3: Forming a decoding coefficient 3 of the camera based on the decoding coefficient 1 and the decoding coefficient 2, and forming an encoding coefficient 3 based on the decoding coefficient 3;

[0012] Step 4: When the image controller receives the product image information transmitted by the camera, it activates encoding coefficient 1, encoding coefficient 2 and encoding coefficient 3 in sequence, and determines whether to process the received product image information based on decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3 in sequence.

[0013] Preferably, in step 1, once the camera is installed, its orientation value is constant. During the normal operation of the camera, the orientation value is refreshed by the operator himself. After the orientation value of the camera changes, the refreshed orientation value is automatically transmitted to the image controller, and then the decoding coefficient 1 is re-formed, and the re-formed decoding coefficient 1 is used to re-form the corresponding encoding coefficient 1.

[0014] Preferably, in step 1, after the two cameras are initially installed and put into operation, the position values ​​of the cameras are also refreshed regularly.

[0015] Preferably, the orientation value may be a coordinate value of the location of the camera located by a GPS module.

[0016] Preferably, in step 1, after the camera is initially installed, if the change address command is not received, the position value of the camera is obtained at each set time interval. If the change address command is received, after the location of the camera is changed, the position value of the camera after the change is obtained, and the position value of the camera is obtained at each set time interval; the set time interval one is lower than the set time interval two.

[0017] Preferably, decoding coefficient one, decoding coefficient two and decoding coefficient three are all stored in information library one of the camera, encoding coefficient one is stored in information library two of the image controller, encoding coefficient two is stored in information library three of the image controller, and encoding coefficient three is stored in information library four of the image controller.

[0018] Preferably, when the image controller receives the product image information transmitted by the camera, the product image information is first stored in the pipeline PIPE 1. When the received product image information is identified and processed, the product image information is sent to the pipeline PIPE 2 for processing. When the processed product image information is received, the identification code of the product image information is extracted, and the product image information received in the pipeline PIPE 1 is removed. The identification code is synchronously stored in the information library 5. The identification codes stored in the information library 5 form alarm messages one by one according to the order of storage, and are transmitted to the screen connected to the image controller.

[0019] Preferably, when the image controller receives product image information transmitted by the camera, the encoding coefficient 1, the encoding coefficient 2, and the encoding coefficient 3 are activated one by one, and whether to process the received product image information is determined based on the decoding coefficient 1, the decoding coefficient 2, and the decoding coefficient 3 one by one, specifically including:

[0020] When the image controller receives the product image information transmitted by the camera, it obtains the camera's position value in the product image information, searches the information base 2, and determines whether there is a coding coefficient 1 corresponding to the position value. If so, it transmits a confirmation command 1 to the camera. After the camera receives the confirmation command 1, it transmits a processing command 1 to the image controller. If not, it transmits a non-processing command to the image controller.

[0021] When the image controller receives the first processing command, the bit rate threshold value when the camera transmits the product image information to the image controller is obtained, and the information database is searched to determine whether the coding coefficient corresponding to the bit rate threshold value exists. If so, a second confirmation command is sent to the camera. After the camera receives the second confirmation command, the second processing command is sent to the image controller. If not, a non-processing command is sent to the image controller.

[0022] When the image controller receives the second processing command, it searches the fourth information base and determines whether there is a coding coefficient third corresponding to the first coding coefficient and the second coding coefficient. If so, it transmits a confirmation command third to the camera. After the camera receives the confirmation command third, it searches the first information base and determines whether there are a decoding coefficient first, a decoding coefficient second, and a decoding coefficient third corresponding to the first coding coefficient, the second coding coefficient, and the third coding coefficient, respectively. If so, it encapsulates the decoding coefficient first, the decoding coefficient second, and the decoding coefficient third and transmits the third processing command to the image controller. If not, it transmits a non-processing command to the image controller.

[0023] After the image controller receives the processing command three, it determines the product image information received for processing and uses the packaged decoding coefficient one, decoding coefficient two and decoding coefficient three to process the received product image information.

[0024] Preferably, obtaining the bit rate threshold when the camera transmits the product image information to the image controller may further include: obtaining a real-time bit rate value when the camera transmits the product image information to the image controller, forming a decoding coefficient 4 according to the real-time bit rate value, and forming an encoding coefficient 4 according to the decoding coefficient 4;

[0025] When the image controller receives the product image information transmitted by the camera, it activates encoding coefficient one, encoding coefficient two, encoding coefficient three and encoding coefficient four in sequence, and determines whether to process the received product image information based on decoding coefficient one, decoding coefficient two, decoding coefficient three and decoding coefficient four in sequence.

[0026] A device used in a machine vision method, comprising:

[0027] a camera for capturing product images and an image controller for processing the product images, the camera and the image controller being connected;

[0028] The device used in the machine vision method also includes:

[0029] A first forming module, which is used to obtain a camera position value, and to form a first decoding coefficient according to the camera position value, and to form a first encoding coefficient according to the first decoding coefficient;

[0030] forming a second module for obtaining a bit rate threshold when the camera transmits the product image to the image controller, forming a second decoding coefficient according to the bit rate threshold, and forming a second encoding coefficient according to the second decoding coefficient;

[0031] A forming module 3 is configured to form a decoding coefficient 3 of the camera according to the decoding coefficient 1 and the decoding coefficient 2, and to form an encoding coefficient 3 according to the decoding coefficient 3;

[0032] The identification module is used to successively enable encoding coefficient 1, encoding coefficient 2 and encoding coefficient 3 when the image controller receives product image information transmitted by the camera, and successively determine whether to process the received product image information based on decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3.

[0033] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0034] Through the above-mentioned machine vision method, the orientation value of the camera is initially obtained, and a decoding coefficient one is formed according to the orientation value of the camera, and an encoding coefficient one is formed according to the decoding coefficient one; then, the bit rate critical value when the camera transmits product image information to the image controller is obtained, and a decoding coefficient two is formed according to the bit rate critical value, and an encoding coefficient two is formed according to the decoding coefficient two; then, a decoding coefficient three of the camera is formed according to the decoding coefficient one and the decoding coefficient two, and an encoding coefficient three is formed according to the decoding coefficient three; finally, when the image controller receives the product image information transmitted by the camera, the encoding coefficient one, the encoding coefficient two and the encoding coefficient three are successively activated, and the decoding coefficient one, the encoding coefficient two and the encoding coefficient three are successively activated according to the decoding coefficient one, the decoding coefficient two and the encoding coefficient three. Decoding coefficient 2 and decoding coefficient 3 determine whether to process the received product image information. Regarding the use of encoding and decoding coefficients, decoding coefficient 1 is formed based on the camera's orientation value, and decoding coefficient 2 is formed based on the bit rate threshold. In addition, decoding coefficient 3 must be formed together with decoding coefficient 1 and decoding coefficient 2. Therefore, when the image controller is performing the rationality of the image information source later, it must ensure that the corresponding encoding coefficient 1, encoding coefficient 2 and encoding coefficient 3 are all reasonable before enabling decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3 to process the received product image information. This can greatly improve the reliability of the image controller in processing product image information. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of the machine vision method described in the present invention;

[0036] Figure 2 It is a partial structural diagram of the device used in the machine vision method described in the present invention. DETAILED DESCRIPTION

[0037] To make the objectives, technical solutions, and advantages of the present invention more clear, the following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely express the technical solutions of the present invention. The embodiments expressed in this application are only some embodiments of the present invention, not all embodiments. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0038] like Figure 1 As shown, a machine vision method according to the present invention includes:

[0039] The camera transmits the product image it captures to the image controller for processing and storage, which specifically includes:

[0040] Step 1: Obtain the camera's position value, generate a decoding coefficient 1 based on the camera's position value, and generate an encoding coefficient 1 based on the decoding coefficient 1;

[0041] Step 2: Obtaining a bit rate threshold when the camera transmits the product image to the image controller, generating a decoding coefficient 2 according to the bit rate threshold, and generating an encoding coefficient 2 according to the decoding coefficient 2;

[0042] Step 3: Forming a decoding coefficient 3 of the camera based on the decoding coefficient 1 and the decoding coefficient 2, and forming an encoding coefficient 3 based on the decoding coefficient 3;

[0043] Step 4: When the image controller receives the product image information transmitted by the camera, it activates encoding coefficient 1, encoding coefficient 2 and encoding coefficient 3 in sequence, and determines whether to process the received product image information based on decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3 in sequence.

[0044] In a preferred but non-restrictive embodiment of the present invention, in step 1, as soon as the camera is installed, its orientation value is constant. During the normal operation of the camera, the refresh of its orientation value is done by the operator himself, just as the operator changes the location of a camera and changes the camera from capturing user electricity usage data of number one to capturing product image information of number two. In this case, after the orientation value of the camera changes, the refreshed orientation value will be automatically transmitted to the image controller, and then the decoding coefficient one will be re-formed, and the re-formed decoding coefficient one will be used to re-form the corresponding encoding coefficient one.

[0045] In a preferred but non-restrictive embodiment of the present invention, in step 1, after the two cameras are initially installed and used, no address change command is received, and no operator changes the camera's position value. After the camera is initially installed, due to different installation locations, the position often changes with the use status. Therefore, in such a situation, the camera's position value must also be refreshed regularly.

[0046] In a preferred but non-limiting embodiment of the present invention, the orientation value may be a coordinate value of the location of the camera located by a GPS module.

[0047] In a preferred but non-restrictive embodiment of the present invention, in step 1, therefore, in this application, after the camera is initially installed, if the address change command is not received, the position value of the camera is obtained at each set time interval. If the address change command is received, after the location of the camera is changed, the position value of the camera after the change is obtained, and the position value of the camera is obtained at each set time interval. The set time interval one is lower than the set time interval two.

[0048] If the camera's address changes due to receiving a change address command, the change is made by the operator. In such cases, the camera's address is often no longer optimal for capturing product images or transmitting product image information to the image controller. Furthermore, the operator often wishes to modify the product being captured by the camera. Under these circumstances, if a hacker impersonates the camera and sends an illegal command to the image controller, they can easily spoof decoding coefficient 1 based on the camera's position change. Therefore, duration 1 should not be set too high and can be set as required, such as 600 minutes, 400 minutes, or 300 minutes. If the camera does not receive a change address command after initial installation, the probability of a hacker spoofing the camera's position is relatively low, as the camera itself does not actively perform address changes, resulting in rare position updates. Therefore, duration 2 corresponds to duration 1 and can be set to a larger value, such as 800 minutes, 900 minutes, or 1000 minutes.

[0049] When the camera's position value changes, the decoding coefficient one and the decoding coefficient one must be refreshed to ensure that the image controller spends less time verifying and receiving product image information. Therefore, in this application, after the camera's position value changes, the decoding coefficient one is re-generated based on the changed camera's position value, and the encoding coefficient one is re-generated based on the re-generated decoding coefficient one.

[0050] However, because after the camera is installed and has captured product images, if the bit rate threshold when the camera transmits product image information to the image controller is kept constant, if there is an illegal attack command to manipulate the camera to transmit product image information with a high bit rate to the image controller, the camera can detect it. Therefore, in this application, the bit rate threshold when the camera transmits product image information to the image controller cannot be changed after the camera is installed. If the number and product range of the products captured by the camera are to be changed, an independent permission command is required. This independent permission command must be executed in the background locally on the camera, and remote operation is usually not allowed. Remote operation is only allowed under the condition that independent command reliability assurance measures are set.

[0051] In a preferred but non-restrictive embodiment of the present invention, all decoding coefficients are stored in the camera after they are formed, while the encoding coefficients are stored in the image controller to achieve reasonable communication between the camera and the image controller. In this application, because command interaction between the camera and the image controller is to be executed, the decoding coefficient one, decoding coefficient two and decoding coefficient three described in this application are all stored in the information library one of the camera, the encoding coefficient one is stored in the information library two of the image controller, the encoding coefficient two is stored in the information library three of the image controller, and the encoding coefficient three is stored in the information library four of the image controller.

[0052] In a preferred but non-limiting embodiment of the present invention, since the image controller has to receive a large amount of product image information, and in many cases the information is received simultaneously, if the received image information is directly stored in a fixed storage interval and then a reliability check is performed on the received image information, then after it is found that the source of the image information is unreliable, the image information will be cleared, which will occupy a lot of storage processes, making the image controller's processing load too heavy. Therefore, in this application, when the image controller receives the product image information transmitted by the camera, the product image information is first stored in pipe PIPE 1. When the received product image information is identified and processed, the product image information is sent to pipe PIPE 2 for processing. When the processed product image information is received, an identification code (a unique symbol set to identify the product image information) of the product image information is extracted, and the product image information received in pipe PIPE 1 is removed. The identification code is synchronously stored in information library 5. The identification codes stored in information library 5 are sequentially generated into alarm messages in the order in which they are stored, and are transmitted to the screen connected to the image controller.

[0053] In a preferred but non-limiting embodiment of the present invention, to ensure the reliability and rationality of the product image information received by the image controller, the present application performs several layers of coding coefficient and decoding coefficient verification. Therefore, in the present application, when the image controller receives the product image information transmitted by the camera, the coding coefficient 1, coding coefficient 2, and coding coefficient 3 are successively activated, and the decoding coefficient 1, decoding coefficient 2, and decoding coefficient 3 are successively used to determine whether to process the received product image information. Specifically, the following steps are performed:

[0054] When the image controller receives the product image information transmitted by the camera, it obtains the camera's position value in the product image information, searches the information base 2, and determines whether there is a coding coefficient 1 corresponding to the position value. If so, it transmits a confirmation command 1 to the camera. After the camera receives the confirmation command 1, it transmits a processing command 1 to the image controller. If not, it transmits a non-processing command to the image controller.

[0055] When the image controller receives the first processing command, the bit rate threshold value when the camera transmits the product image information to the image controller is obtained, and the information database is searched to determine whether the coding coefficient corresponding to the bit rate threshold value exists. If so, a second confirmation command is sent to the camera. After the camera receives the second confirmation command, the second processing command is sent to the image controller. If not, a non-processing command is sent to the image controller.

[0056] When the image controller receives the second processing command, it searches the fourth information base and determines whether there is a coding coefficient third corresponding to the first coding coefficient and the second coding coefficient. If so, it transmits a confirmation command third to the camera. After the camera receives the confirmation command third, it searches the first information base and determines whether there are a decoding coefficient first, a decoding coefficient second, and a decoding coefficient third corresponding to the first coding coefficient, the second coding coefficient, and the third coding coefficient, respectively. If so, it encapsulates the decoding coefficient first, the decoding coefficient second, and the decoding coefficient third and transmits the third processing command to the image controller. If not, it transmits a non-processing command to the image controller.

[0057] After receiving the processing command 3, the image controller determines that the received product image information is processed and processes the received product image information using the encapsulated decoding coefficients 1, 2, and 3. A "do not process" command does not process the product image information, while a "process" command processes the product image information.

[0058] When using the decoding coefficient to process the product image information, the image controller only uses the decoding coefficient as a one-time coefficient. That is, when using the encapsulated decoding coefficient 1, decoding coefficient 2, and decoding coefficient 3 to process the received product image information, after processing, the encapsulated decoding coefficient 1, decoding coefficient 2, and decoding coefficient 3 are immediately removed, or the encapsulated decoding coefficient 1, decoding coefficient 2, and decoding coefficient 3 are set to a dormant state to prevent the product image information that has not yet been verified in the image controller from stealing the decoding coefficient, thereby forcing the image controller to reasonably process the product image information.

[0059] In a preferred but non-limiting embodiment of the present invention, even if a camera is associated with a corresponding product, there may be situations where the product is not captured for an extended period. Therefore, in such situations, it is only necessary to obtain the real-time bit rate value of the product image information transmitted by the camera to generate a corresponding decoding coefficient. The decoding coefficient is then used to simultaneously generate a corresponding encoding coefficient. The generated encoding coefficient and decoding coefficient can also serve as a basis for determining whether the image controller executes product image information processing. Therefore, in this application, obtaining the bit rate threshold value when the camera transmits product image information to the image controller may further include: obtaining the real-time bit rate value when the camera transmits product image information to the image controller, generating decoding coefficient 4 based on the real-time bit rate value, and generating encoding coefficient 4 based on decoding coefficient 4;

[0060] When the image controller receives product image information transmitted by the camera, it sequentially activates encoding coefficients 1, 2, 3, and 4. It then determines whether to process the received product image information based on decoding coefficients 1, 2, 3, and 4. The encoding and decoding coefficients can be entropy encoding or decoding, respectively. The basis for forming the encoding coefficients or decoding coefficients is the basis for using the encoding coefficients or decoding coefficients as the encoding coefficients or decoding coefficients.

[0061] like Figure 2 As shown, the device used in the machine vision method described in the present invention includes:

[0062] A camera for capturing product images and an image controller for processing product images, the camera and the image controller being connected, and the camera transmitting the captured product images to the image controller for processing and storage;

[0063] The device used in the machine vision method also includes:

[0064] A first forming module, which is used to obtain a camera position value, and to form a first decoding coefficient according to the camera position value, and to form a first encoding coefficient according to the first decoding coefficient;

[0065] forming a second module for obtaining a bit rate threshold when the camera transmits the product image to the image controller, forming a second decoding coefficient according to the bit rate threshold, and forming a second encoding coefficient according to the second decoding coefficient;

[0066] A forming module 3 is configured to form a decoding coefficient 3 of the camera according to the decoding coefficient 1 and the decoding coefficient 2, and to form an encoding coefficient 3 according to the decoding coefficient 3;

[0067] The identification module is used to successively enable encoding coefficient 1, encoding coefficient 2 and encoding coefficient 3 when the image controller receives the product image information transmitted by the camera, and successively determine whether to process the received product image information based on decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3.

[0068] The beneficial effects of the present invention are as follows:

[0069] Through the above-mentioned machine vision method, the orientation value of the camera is initially obtained, and a decoding coefficient one is formed according to the orientation value of the camera, and an encoding coefficient one is formed according to the decoding coefficient one; then, the bit rate critical value when the camera transmits product image information to the image controller is obtained, and a decoding coefficient two is formed according to the bit rate critical value, and an encoding coefficient two is formed according to the decoding coefficient two; then, a decoding coefficient three of the camera is formed according to the decoding coefficient one and the decoding coefficient two, and an encoding coefficient three is formed according to the decoding coefficient three; finally, when the image controller receives the product image information transmitted by the camera, the encoding coefficient one, the encoding coefficient two and the encoding coefficient three are successively activated, and the decoding coefficient one, the encoding coefficient two and the encoding coefficient three are successively activated according to the decoding coefficient one, the decoding coefficient two and the encoding coefficient three. Decoding coefficient 2 and decoding coefficient 3 determine whether to process the received product image information. Regarding the use of encoding and decoding coefficients, decoding coefficient 1 is formed based on the camera's orientation value, and decoding coefficient 2 is formed based on the bit rate threshold. In addition, decoding coefficient 3 must be formed together with decoding coefficient 1 and decoding coefficient 2. Therefore, when the image controller is performing the rationality of the image information source later, it must ensure that the corresponding encoding coefficient 1, encoding coefficient 2 and encoding coefficient 3 are all reasonable before enabling decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3 to process the received product image information. This can greatly improve the reliability of the image controller in processing product image information.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not deviate from the spirit and scope of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. A machine vision method, characterized in that: include: The camera transmits the captured product image to the image controller for processing, which includes: Step 1: Obtain the position value of the camera, and generate a decoding coefficient 1 according to the position value of the camera, and generate an encoding coefficient 1 according to the decoding coefficient 1; Step 2: obtaining a bit rate threshold when the camera transmits the product image to the image controller, and forming a decoding coefficient 2 according to the bit rate threshold, and forming an encoding coefficient 2 according to the decoding coefficient 2; Step 3: forming a decoding coefficient 3 of the camera according to the decoding coefficient 1 and the decoding coefficient 2, and forming an encoding coefficient 3 according to the decoding coefficient 3; Step 4: When the image controller receives the product image information transmitted by the camera, the encoding coefficient 1, the encoding coefficient 2 and the encoding coefficient 3 are activated one by one, and it is determined whether to process the received product image information according to the decoding coefficient 1, the decoding coefficient 2 and the decoding coefficient 3 one by one.

2. The machine vision method according to claim 1, characterized in that: In step 1, once the camera is installed, its position value is constant. During the normal operation of the camera, its position value is refreshed in two ways: one is refreshed by the operator. After the position value of the camera changes, the refreshed position value is automatically transmitted to the image controller, and then a decoding coefficient 1 is generated again, and the corresponding encoding coefficient 1 is generated again using the decoder coefficient 1 generated again.

3. The machine vision method according to claim 2, characterized in that: In step 1, after the two cameras are initially installed and put into use, the camera position values ​​are also updated regularly.

4. The machine vision method according to claim 3, characterized in that: In step 1, the position value may be a coordinate value of the location of the camera located by a GPS module.

5. The machine vision method according to claim 4, characterized in that: In step 1, after the camera is initially installed, if no change address command is received, the position value of the camera is obtained at each set time interval of one. If a change address command is received, after the location of the camera is changed, the position value of the camera after the change is obtained, and the position value of the camera is obtained at each set time interval of two; the set time interval one is shorter than the set time interval two.

6. The machine vision method according to claim 5, characterized in that: Decoding coefficient one, decoding coefficient two and decoding coefficient three are all stored in information base one of the camera, encoding coefficient one is stored in information base two of the image controller, encoding coefficient two is stored in information base three of the image controller, and encoding coefficient three is stored in information base four of the image controller.

7. The machine vision method according to claim 6, characterized in that: When the image controller receives the product image information transmitted by the camera, the product image information is first stored in the pipeline PIPE 1. When the received product image information is identified and processed, the product image information is sent to the pipeline PIPE 2 for processing. When the processed product image information is received, the identification code of the product image information is extracted, and the product image information received in the pipeline PIPE 1 is removed. The identification code is synchronously stored in the information library 5. The identification codes stored in the information library 5 form alarm messages one by one according to the order of storage, and are transmitted to the screen connected to the image controller.

8. The machine vision method according to claim 7, characterized in that: When the image controller receives the product image information transmitted by the camera, the encoding coefficient 1, the encoding coefficient 2 and the encoding coefficient 3 are activated one by one, and the decoding coefficient 1, the decoding coefficient 2 and the decoding coefficient 3 are used to determine whether to process the received product image information, including: When the image controller receives the product image information transmitted by the camera, it obtains the camera position value in the product image information, searches the information base 2, and determines whether there is a coding coefficient 1 corresponding to the position value. If so, it transmits a confirmation command 1 to the camera. After the camera receives the confirmation command 1, it transmits a processing command 1 to the image controller. If not, it transmits a non-processing command to the image controller. When the image controller receives the processing command 1, the bit rate threshold when the camera transmits the product image information to the image controller is obtained, and the information base 3 is searched to determine whether there is a coding coefficient 2 corresponding to the bit rate threshold. If there is, a confirmation command 2 is transmitted to the camera. After the camera receives the confirmation command 2, the processing command 2 is transmitted to the image controller. If not, a non-processing command is transmitted to the image controller. When the image controller receives the processing command 2, it searches the information base 4 and determines whether there is a coding coefficient 3 corresponding to the coding coefficient 1 and the coding coefficient 2. If yes, it transmits the confirmation command 3 to the camera. After the camera receives the confirmation command 3, it searches the information base 1 and determines whether there are decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3 corresponding to the coding coefficient 1, coding coefficient 2 and coding coefficient 3 respectively. If yes, it encapsulates the decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3 and transmits the processing command 3 to the image controller. If no, it transmits the non-processing command to the image controller. After the image controller receives the processing command three, it determines and processes the received product image information, and uses the packaged decoding coefficient one, decoding coefficient two and decoding coefficient three to process the received product image information.

9. The machine vision method according to claim 8, characterized in that: When obtaining the bit rate threshold when the camera transmits the product image information to the image controller, the method may further include: obtaining the instantaneous bit rate value when the camera transmits the product image information to the image controller, and forming a decoding coefficient 4 according to the instantaneous bit rate value, and forming an encoding coefficient 4 according to the decoding coefficient 4; When the image controller receives the product image information transmitted by the camera, the encoding coefficient 1, encoding coefficient 2, encoding coefficient 3 and encoding coefficient 4 are activated one by one, and it is determined whether to process the received product image information according to the decoding coefficient 1, decoding coefficient 2, decoding coefficient 3 and decoding coefficient 4 one by one.

10. A device for a machine vision method, characterized in that: include: a camera for capturing an image of a product and an image controller for processing the image of the product, the camera and the image controller being connected; The device used in the machine vision method also includes: A forming module 1, which is used to obtain the position value of the camera, and to form a decoding coefficient 1 according to the position value of the camera, and to form an encoding coefficient 1 according to the decoding coefficient 1; Forming module 2, which is used to obtain a bit rate threshold when the camera transmits the product image to the image controller, and to form a decoding coefficient 2 according to the bit rate threshold, and to form a coding coefficient 2 according to the decoding coefficient 2; A forming module 3 is used to form a decoding coefficient 3 of the camera according to the decoding coefficient 1 and the decoding coefficient 2, and to form an encoding coefficient 3 according to the decoding coefficient 3; The identification module is used to successively enable encoding coefficient 1, encoding coefficient 2 and encoding coefficient 3 when the image controller receives the product image information transmitted by the camera, and successively identify whether to process the received product image information based on decoding coefficient 1, decoding coefficient 2 and decoding coefficient 3.

Images