Decomposition cutting path planning method in processing of anchovy
Patent Information
- Application Number
- CN202610857075.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-15
- Publication Date
- 2026-09-25
AI Technical Summary
[0003]现有鳀鱼切割多采用固定路径切割模式，即预设统一的切割角度和路径，忽略鳀鱼个体位姿差异（如身体倾斜、俯仰）及鱼头位姿变化（如鱼头偏转、俯仰），导致切割过程中易出现鱼头切割不彻底、内脏残留、鱼肉过度损耗等问题，严重影响产品品质与加工效率
1.在切割的时候实时获取鱼体中轴线和X轴的夹角和鱼头中轴线与鱼体中轴线的夹角，通过对应两个夹角实时调整切割角度，有效解决鱼头切割不彻底、内脏残留和鱼肉过度损耗等问题，进而提升产品品质；
Smart Images
Figure CN122820822A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of food processing technology, and more specifically, to a method for planning the decomposition and cutting path during anchovy processing. Background Technology
[0002] Anchovies, small, mid-to-upper-level migratory fish, have a laterally compressed body, no lateral line, and a cylindrical abdomen. They typically measure 8-12 cm in length. Their flesh is tender and nutritious, making them an important raw material in the aquatic product processing industry, widely used in the production of fish paste, canned fish, and animal feed. One of the core steps in anchovy processing is head and viscera removal and segmentation. The precision of this cutting directly affects the utilization rate of the fish meat, product quality, and subsequent processing efficiency. Due to the anchovy's small size, irregular shape, and significant individual variation, and the ease with which its position can shift during processing due to transport and clamping, coupled with its fine muscle fibers and the tendency for a water film to form on its surface, traditional cutting methods have many drawbacks.
[0003] Current anchovy cutting methods mostly employ fixed-path cutting, which pre-sets a uniform cutting angle and path, ignoring individual anchovy pose differences (such as body tilt and pitch) and head pose changes (such as head rotation and pitch). This leads to problems such as incomplete head cutting, visceral residue, and excessive meat loss during the cutting process, severely impacting product quality and processing efficiency. Some improved technologies attempt to use simple machine vision for positioning, but these rely heavily on traditional convolutional neural networks for image processing. These methods suffer from incomplete feature extraction, insensitivity to subtle pose changes in anchovies, and weak anti-interference capabilities. They cannot accurately capture the feature differences caused by the anchovy's irregular body shape and surface water film reflection, making accurate pose recognition difficult. Summary of the Invention
[0004] In order to solve at least one of the above-mentioned technical problems, the present invention aims to provide a method for planning the decomposition and cutting path during anchovy processing, which can improve the efficiency of anchovy viscera removal.
[0005] This invention provides a method for planning the decomposition and cutting path during anchovy processing, comprising: Obtain multi-dimensional image data of a fixed anchovy, and construct a three-dimensional geometric coordinate model of the anchovy based on the multi-dimensional image data; Based on the three-dimensional geometric coordinate model of the anchovy, the anchovy's pose parameters and the direction angle of the fish meat fibers are determined. The pose parameters include the angle between the fish's central axis and the horizontal direction, and the angle between the fish's head central axis and the fish's body central axis. The angle of the clamping knife is obtained based on the angle between the fish's central axis and the horizontal direction, the angle between the fish's head central axis and the fish's body central axis, and the direction angle of the fish meat fibers. Based on the fixed three-dimensional geometric coordinate model of the anchovy, the connection surface between the head and body and the central axis of the body are determined; Using the angled cutting angle of the clamping knife as a reference, first cut along the connection surface between the fish head and the fish body, then cut along the central axis of the fish body until the cut is complete.
[0006] In this solution, after obtaining the multi-dimensional image data information of the fixed anchovies, the solution further includes: Convert the color RGB image of anchovies in multi-dimensional image data to a grayscale image; Within the preset ROI area, the grayscale distribution is statistically analyzed pixel by pixel to determine the contrast characteristic parameters of each area; If the contrast feature parameters of a region are lower than the minimum value in the preset contrast range, then the contrast feature parameters of the corresponding region need to be equalized and enhanced. If the contrast feature parameters of a region are within a preset contrast range, then the contrast feature parameters of the corresponding region need to be slightly equalized. If the contrast characteristic parameters of a region are higher than the preset contrast range, then the contrast characteristic parameters of the corresponding region do not need to be processed.
[0007] In this solution, the step of determining the contrast characteristic parameters of each region specifically includes: Count the number of pixels corresponding to each gray level and the total number within the preset ROI area; The average gray level of the corresponding region is determined based on the number of pixels corresponding to each gray level and the total number of pixels within the preset ROI region. Calculate the difference between the gray levels of any two pixels in the preset ROI region to determine the gray level difference; after traversing all pixels, obtain the set of gray level differences. Extract the maximum value D from the set of grayscale differences; Based on the average gray level and the maximum value D in the set of gray level differences for the corresponding region, the contrast characteristic parameter C of the corresponding region is determined, and the formula is as follows: ,in The target gray level for the corresponding area. This represents the average gray level of the corresponding region.
[0008] In this solution, the equalization enhancement step specifically includes: Divide the corresponding preset ROI region into The sub-block; The cumulative distribution function (CDF) is calculated independently for each sub-block, and its formula is as follows: ,in Let be the gray level probability density corresponding to gray level i; The contrast enhancement value for each sub-block is determined based on the cumulative integral function of each sub-block. Its formula is .
[0009] In this solution, the step of constructing the three-dimensional geometric coordinate model of the anchovy specifically includes: Based on multi-dimensional image data of anchovies, a three-dimensional geometric model was constructed, including the head, body, tail, dorsal fin, pelvic fin, vertebrae, and abdominal cavity. Using the foremost point of the fish head in the 3D geometric model as the origin O, the fish's central axis overlaps with the preset X-axis and the X-axis points towards the fish tail, the Y-axis is perpendicular to the X-axis and points to the left side of the fish body, and the Z-axis is perpendicular to the X-axis and points upwards, thus constructing a 3D geometric coordinate model.
[0010] This plan also includes: When cutting along the connection surface between the fish head and the fish body, the angle between the central axis of the fish head and the central axis of the fish body is obtained in real time. Sequentially adjust the angle between the fish head's central axis and the fish body's central axis, and the preset angle. Perform the difference calculation to obtain the first included angle difference; If the absolute value of the first included angle difference is greater than the preset first included angle threshold, then the first included angle difference is multiplied by the corresponding correction coefficient to obtain the first oblique angle correction value. Add the first chamfer angle correction value to the chamfer angle to obtain the corrected chamfer angle; Continue cutting the fish head using the corrected angled cut.
[0011] This plan also includes: When cutting along the midline of the fish, the angle between the midline of the fish and the X-axis is obtained in real time; Sequentially adjust the angle between the fish's central axis and the X-axis to the preset angle. Perform the difference calculation to obtain the second included angle difference; If the absolute value of the second included angle difference is greater than the preset second included angle threshold, then the second included angle difference is multiplied by the corresponding correction coefficient to obtain the second oblique angle correction value. Add the second chamfer angle correction value to the chamfer angle to obtain the corrected chamfer angle; Continue cutting the fish body at the corrected angle.
[0012] This plan also includes: When cutting along the central axis of the fish, the distance from the current cutting position to the end of the central axis of the fish is obtained in real time; If the distance from the current cutting position to the end of the fish's central axis is less than or equal to a preset distance threshold, rotate the preset cutting angle. Adjust the cutting angle of the clamping blade according to the preset cutting angle to rotate and cut the fish until the internal organs and the fish body are separated.
[0013] In this solution, the step of determining the connection surface between the fish head and the fish body specifically includes: Extract the following points from the three-dimensional geometric coordinate model of the anchovy: C1 (midpoint of the posterior margin of the skull), C2 (lowest point of the posterior margin of the left operculum), C3 (lowest point of the posterior margin of the right operculum), and C4 (lowest point on the dorsal side at the junction of the occipital condyle and the first vertebra). Using four feature points C1, C2, C3, and C4 as control points, a connection surface is generated by fitting a pre-defined triple bilinear B-spline surface. ,in These are parameters along the width of the fish's body. These are parameters along the height of the fish's body.
[0014] One or more technical solutions proposed in this application have at least the following technical effects: 1. During cutting, the angle between the fish body's central axis and the X-axis, and the angle between the fish head's central axis and the fish body's central axis are obtained in real time. The cutting angle is adjusted in real time based on the two corresponding angles, which effectively solves problems such as incomplete cutting of the fish head, internal organ residue, and excessive loss of fish meat, thereby improving product quality. 2. A customized solution was designed for anchovies, which are small in size, vary greatly in size, are prone to shifting in position, and are prone to forming a water film on their surface. It does not rely on a fixed path and can be adapted to anchovies of different sizes and freshness. At the same time, it is suitable for the needs of continuous industrial production and solves the problem of poor adaptability of existing technologies to anchovies. 3. The planned two-stage decomposition cutting path enables the simultaneous completion of head removal and viscera removal. Combined with the synchronous control of cutting and conveying, it improves processing efficiency, reduces secondary processing steps, and lowers labor costs. 4. It adopts a conventional depth camera vision system and clamping structure, which is easy to modify and has controllable cost. It can be quickly applied to existing anchovy processing production lines. At the same time, it optimizes the cutting angle by combining the direction of anchovy muscle fibers, reduces fish meat fragmentation, and further improves fish meat utilization, which has extremely high industrial application value. Attached Figure Description
[0015] The present invention will now be described in further detail with reference to the accompanying drawings and specific implementation methods.
[0016] Figure 1 A flowchart of a method for decomposition and cutting path planning in anchovy processing according to the present invention is shown. Detailed Implementation
[0017] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.
[0018] Figure 1 A flowchart of a method for decomposition and cutting path planning in anchovy processing according to the present invention is shown.
[0019] like Figure 1 As shown, this invention discloses a method for decomposition and cutting path planning during anchovy processing, including: S101, Obtain multi-dimensional image data information of a fixed anchovy, and construct a three-dimensional geometric coordinate model of the anchovy based on the multi-dimensional image data; S102, Based on the three-dimensional geometric coordinate model of the anchovy, determine the anchovy's pose parameters and the direction angle of the fish meat fibers. The pose parameters include the angle between the fish's central axis and the horizontal direction, and the angle between the fish's head central axis and the fish's body central axis. S103. Based on the angle between the fish's central axis and the horizontal direction, the angle between the fish's head central axis and the fish's central axis, and the direction angle of the fish meat fibers, the oblique cutting angle of the clamping knife is obtained. S104, Based on the three-dimensional geometric coordinate model of the anchovy, determine the connection surface between the head and body and the central axis of the body; S105, using the angled cutting angle of the clamping knife as a reference, first cut along the connection surface between the fish head and the fish body, then cut along the central axis of the fish body until the cut is complete.
[0020] According to an embodiment of the present invention, the anchovy is first fixed to obtain a fixed anchovy, and then the fixed anchovy is photographed from multiple angles to obtain multi-dimensional image data. The angle of the clamping knife is set to... Its formula is ,in As the reference angle, , These are the correction coefficients for the fish body tilt angle and the fish head deflection angle, respectively, where 'a' is the angle between the fish body's central axis and the horizontal direction (fish body tilt angle). The angle between the central axis of the fish's head and the central axis of the fish's body (fish head deflection angle). The angle of fish meat fiber direction.
[0021] Further, the step of obtaining the fish meat fiber direction angle is as follows: select the complete trunk meat area from the gill cover dividing point to the middle section of the fish body as the muscle extraction area, and use the fish body central axis as a reference to calculate the angle between the direction of the fish meat fiber and the fish body central axis, which is the fish meat fiber direction angle.
[0022] Furthermore, the steps for obtaining the fish's central axis are as follows: select the tip of the anchovy's tail fin as the tail vertex, select the midpoint of the posterior edge of the anchovy's gill cover as the gill cover boundary point, connect the corresponding two points and fit a straight line in space using the least squares method, and the resulting straight line is the fish's central axis.
[0023] Furthermore, the steps for obtaining the central axis of the fish head are as follows: select the tip of the snout at the front end of the anchovy's mouth as the apex of the snout, use the common boundary point of the posterior edge of the gill cover as the junction base point, connect the apex of the snout and the boundary point of the posterior edge of the gill cover and fit a straight line in space, and the resulting straight line is the central axis of the fish head.
[0024] According to an embodiment of the present invention, after acquiring the multi-dimensional image data information of a fixed anchovy, the method further includes: Convert the color RGB image of anchovies in multi-dimensional image data to a grayscale image; Within the preset ROI area, the grayscale distribution is statistically analyzed pixel by pixel to determine the contrast characteristic parameters of each area; If the contrast feature parameters of a region are lower than the minimum value in the preset contrast range, then the contrast feature parameters of the corresponding region need to be equalized and enhanced. If the contrast feature parameters of a region are within a preset contrast range, then the contrast feature parameters of the corresponding region need to be slightly equalized. If the contrast characteristic parameters of a region are higher than the preset contrast range, then the contrast characteristic parameters of the corresponding region do not need to be processed.
[0025] It should be noted that equalization enhances the fish's body edges and head features, thus facilitating identification. The preset ROI regions include ROI1, ROI2, and ROI3. ROI1 represents the entire fish body and is used to enhance the contrast of the fish's edges. ROI2 represents the fish's head region and is used to enhance the contrast of the eyes and gills. ROI3 represents the fish's central axis region and is used to enhance pose recognition features. For example, the preset contrast range can be set to... If the contrast feature parameter of a region is lower than 0.15, the contrast feature parameter of the corresponding region is enhanced; if the contrast feature parameter of a region is greater than or equal to 0.15 and less than or equal to 0.35, the contrast feature parameter of the corresponding region needs to be slightly equalized, for example, the slight equalization is 1 / 3 of the equalization enhancement.
[0026] According to an embodiment of the present invention, the step of determining the contrast characteristic parameters of each region specifically includes: Count the number of pixels corresponding to each gray level and the total number within the preset ROI area; The average gray level of the corresponding region is determined based on the number of pixels corresponding to each gray level and the total number of pixels within the preset ROI region. Calculate the difference between the gray levels of any two pixels in the preset ROI region to determine the gray level difference; after traversing all pixels, obtain the set of gray level differences. Extract the maximum value D from the set of grayscale differences; Based on the average gray level and the maximum value D in the set of gray level differences for the corresponding region, the contrast characteristic parameter C of the corresponding region is determined, and the formula is as follows: ,in The target gray level for the corresponding area. This represents the average gray level of the corresponding region.
[0027] It should be noted that the method of determining the contrast feature parameters of the entire region by comparing and analyzing the grayscale of each pixel is simple and convenient.
[0028] According to an embodiment of the present invention, the equalization enhancement step specifically includes: Divide the corresponding preset ROI region into The sub-block; The cumulative distribution function (CDF) is calculated independently for each sub-block, and its formula is as follows: ,in Let be the gray level probability density corresponding to gray level i; The contrast enhancement value for each sub-block is determined based on the cumulative integral function of each sub-block. Its formula is .
[0029] It should be noted that after iterating through the contrast enhancement values of all sub-blocks, the contrast of any two adjacent sub-blocks after enhancement is extracted, and the difference is calculated to obtain the contrast difference between adjacent sub-blocks. If the contrast difference between adjacent sub-blocks is greater than the preset contrast threshold, the contrast enhancement values of the corresponding two adjacent sub-blocks are adjusted based on the preset smoothing coefficient. The maximum contrast value in the corresponding adjacent sub-block is multiplied by the preset smoothing coefficient to obtain the adjustment value of the maximum contrast value in the corresponding adjacent sub-block. For example, if the preset smoothing coefficient is 0.8 and the maximum contrast value in the adjacent sub-block is 0.3, then the contrast of the corresponding sub-block after adjustment is 0.8 * 0.3 = 0.24.
[0030] It should be noted that CDF(k) is the cumulative distribution function of gray levels; k is the gray level corresponding to the current calculation; i is the gray level traversal variable; ∑ is the cumulative summation symbol, indicating that it accumulates sequentially from the lowest gray level 0 to gray level k; p(i) is the gray probability density corresponding to gray level i, satisfying... ,in N represents the number of pixels corresponding to the gray level, and N is the total number of pixels in the sub-region.
[0031] According to an embodiment of the present invention, the step of constructing a three-dimensional geometric coordinate model of an anchovy specifically includes: Based on multi-dimensional image data of anchovies, a three-dimensional geometric model was constructed, including the head, body, tail, dorsal fin, pelvic fin, vertebrae, and abdominal cavity. Using the foremost point of the fish head in the 3D geometric model as the origin O, the fish's central axis overlaps with the preset X-axis and the X-axis points towards the fish tail, the Y-axis is perpendicular to the X-axis and points to the left side of the fish body, and the Z-axis is perpendicular to the X-axis and points upwards, thus constructing a 3D geometric coordinate model.
[0032] According to an embodiment of the present invention, it further includes: When cutting along the connection surface between the fish head and the fish body, the angle between the central axis of the fish head and the central axis of the fish body is obtained in real time. Sequentially adjust the angle between the fish head's central axis and the fish body's central axis, and the preset angle. Perform the difference calculation to obtain the first included angle difference; If the absolute value of the first included angle difference is greater than the preset first included angle threshold, then the first included angle difference is multiplied by the corresponding correction coefficient to obtain the first oblique angle correction value. Add the first chamfer angle correction value to the chamfer angle to obtain the corrected chamfer angle; Continue cutting the fish head using the corrected angled cut.
[0033] It should be noted that, for example, if the preset first included angle threshold is 1 degree, then when the difference in the first included angle is equal to or less than 1 degree, the oblique cutting angle of the clamping blade will not be adjusted; the preset included angle This is to correspond to the angle between the fish head's central axis and the fish's body central axis when the angled cutting angle of the clamping knife was last adjusted; for example, if the angle between the fish head's central axis and the fish's body central axis was 12 degrees, and the angled cutting angle of the clamping knife was adjusted, then the preset angle is... It equals 12 degrees; for example, the first included angle difference is 3, and the corresponding correction factor is... The corrected bevel angle increases. At the same time, when the fish head deflects, the rotation direction of the clamping knife is adjusted to be consistent with the direction of the fish head deflection; furthermore, when the first oblique cutting angle correction value is greater than the set correction angle, the corresponding oblique cutting angle is adjusted step by step to improve the smoothness of cutting; when cutting the fish head, the cutting depth is controlled at 1 / 5 to 1 / 4 of the fish body length.
[0034] According to an embodiment of the present invention, it further includes: When cutting along the midline of the fish, the angle between the midline of the fish and the X-axis is obtained in real time; Sequentially adjust the angle between the fish's central axis and the X-axis to the preset angle. Perform the difference calculation to obtain the second included angle difference; If the absolute value of the second included angle difference is greater than the preset second included angle threshold, then the second included angle difference is multiplied by the corresponding correction coefficient to obtain the second oblique angle correction value. Add the second chamfer angle correction value to the chamfer angle to obtain the corrected chamfer angle; Continue cutting the fish body at the corrected angle.
[0035] It should be noted that the preset second included angle threshold is less than or equal to the preset first included angle threshold. The smaller the preset second included angle threshold, the more frequently the angle of the clamping blade is adjusted, resulting in more precise cutting of the fish body. To correspond to the angle between the fish's centerline and the X-axis when the clamping knife last corrected the angled cut, for example, if the angle between the fish's centerline and the X-axis was 6 degrees when the clamping knife last corrected the angled cut, then the preset angle is... Degrees; if the difference in the second included angle is 2, the corresponding correction factor is... The corrected bevel angle increases. At the same time, when the fish body deflects, the rotation direction of the clamping knife is adjusted to be consistent with the direction of the fish body deflection; when cutting the fish body, the clamping knife extends into the fish belly to cut, and the cutting depth is controlled to be 0.8-1.2 cm.
[0036] According to an embodiment of the present invention, it further includes: When cutting along the central axis of the fish, the distance from the current cutting position to the end of the central axis of the fish is obtained in real time; If the distance from the current cutting position to the end of the fish's central axis is less than or equal to a preset distance threshold, rotate the preset cutting angle. Adjust the cutting angle of the clamping blade according to the preset cutting angle to rotate and cut the fish until the internal organs and the fish body are separated.
[0037] It should be noted that by utilizing the rotation function of the clamping knife, when the distance from the end of the fish's central axis is less than or equal to a preset distance threshold, the clamping knife is rotated at a preset cutting angle to ensure that the internal organs are completely separated from the fish body and to avoid internal organ residue. For example, the preset cutting angle is 5 to 10 degrees.
[0038] Furthermore, when the distance from the cutting position to the end of the fish's central axis is less than or equal to a preset distance threshold, the following can also be done: extract the highest point A and the lowest point B of the abdominal cavity from the three-dimensional geometric coordinate model of the anchovy; at the midpoint A1 between points A and B, cut along the negative Z-axis to the bottom wall of the abdominal cavity, and then make a horizontal cut along the positive X-axis to the X coordinate position corresponding to point B, forming an auxiliary incision for visceral dissection, where the X coordinate of point A1 is... The Z-coordinate is .
[0039] According to an embodiment of the present invention, the step of determining the connection surface between the fish head and the fish body specifically includes: Extract the following points from the three-dimensional geometric coordinate model of the anchovy: C1 (midpoint of the posterior margin of the skull), C2 (lowest point of the posterior margin of the left operculum), C3 (lowest point of the posterior margin of the right operculum), and C4 (lowest point on the dorsal side at the junction of the occipital condyle and the first vertebra). Using four feature points C1, C2, C3, and C4 as control points, a connection surface is generated by fitting a pre-defined triple bilinear B-spline surface. ,in These are parameters along the width of the fish's body. These are parameters along the height of the fish's body.
[0040] It should be noted that extracting the connection surface Mid-ridge line As the optimal spatial cutting curve.
[0041] This invention discloses a method for planning the decomposition and cutting path during anchovy processing, comprising: acquiring multi-dimensional image data of a fixed anchovy and constructing a three-dimensional geometric coordinate model of the anchovy based on the multi-dimensional image data; determining the anchovy's pose parameters and the direction angle of the fish meat fibers based on the three-dimensional geometric coordinate model of the anchovy; obtaining the oblique cutting angle of the clamping knife based on the angle between the fish's central axis and the horizontal direction, the angle between the fish's head central axis and the fish's body central axis, and the direction angle of the fish meat fibers; determining the connection surface between the fish's head and body and the fish's central axis based on the fixed three-dimensional geometric coordinate model of the anchovy; using the oblique cutting angle of the clamping knife as a reference, first cutting along the connection surface between the fish's head and body, and then cutting along the fish's body central axis until the cutting is complete; by continuously correcting the cutting path, the cutting accuracy is improved, thereby improving the efficiency of separating the internal organs and the fish body.
[0042] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.
Claims
1. A method for planning the decomposition and cutting path during anchovy processing, characterized in that, include: Obtain multi-dimensional image data of a fixed anchovy, and construct a three-dimensional geometric coordinate model of the anchovy based on the multi-dimensional image data; Based on the three-dimensional geometric coordinate model of the anchovy, the anchovy's pose parameters and the direction angle of the fish meat fibers are determined. The pose parameters include the angle between the fish's central axis and the horizontal direction, and the angle between the fish's head central axis and the fish's body central axis. The angle of the clamping knife is obtained based on the angle between the fish's central axis and the horizontal direction, the angle between the fish's head central axis and the fish's body central axis, and the direction angle of the fish meat fibers. Based on the fixed three-dimensional geometric coordinate model of the anchovy, the connection surface between the head and body and the central axis of the body are determined; Using the angled cutting angle of the clamping knife as a reference, first cut along the connection surface between the fish head and the fish body, then cut along the central axis of the fish body until the cut is complete.
2. The method for decomposition and cutting path planning in anchovy processing according to claim 1, characterized in that, After acquiring the multi-dimensional image data information of the fixed anchovies, the process also includes: Convert the color RGB image of anchovies in multi-dimensional image data to a grayscale image; Within the preset ROI area, the grayscale distribution is statistically analyzed pixel by pixel to determine the contrast characteristic parameters of each area; If the contrast feature parameters of a region are lower than the minimum value in the preset contrast range, then the contrast feature parameters of the corresponding region need to be equalized and enhanced. If the contrast feature parameters of a region are within a preset contrast range, then the contrast feature parameters of the corresponding region need to be slightly equalized. If the contrast characteristic parameters of a region are higher than the preset contrast range, then the contrast characteristic parameters of the corresponding region do not need to be processed.
3. The method for decomposition and cutting path planning in anchovy processing according to claim 2, characterized in that, The step of determining the contrast characteristic parameters of each region specifically includes: Count the number of pixels corresponding to each gray level and the total number within the preset ROI area; The average gray level of the corresponding region is determined based on the number of pixels corresponding to each gray level and the total number of pixels within the preset ROI region. Calculate the difference between the gray levels of any two pixels in the preset ROI region to determine the gray level difference; after traversing all pixels, obtain the set of gray level differences; Extract the maximum value D from the set of grayscale differences; Based on the maximum value D in the set of average gray level and gray level difference of the corresponding region, the contrast feature parameter C of the corresponding region is determined, and the formula is as follows: ,in The target gray level for the corresponding area. This represents the average gray level of the corresponding region.
4. The method for decomposition and cutting path planning in anchovy processing according to claim 2, characterized in that, The equalization enhancement steps specifically include: Divide the corresponding preset ROI region into The sub-block; The cumulative distribution function (CDF) is calculated independently for each sub-block, and its formula is as follows: ,in Let be the gray level probability density corresponding to gray level i; The contrast enhancement value for each sub-block is determined based on the cumulative integral function of each sub-block. Its formula is .
5. The method for decomposition and cutting path planning in anchovy processing according to claim 1, characterized in that, The steps for constructing the three-dimensional geometric coordinate model of the anchovy specifically include: Based on multi-dimensional image data of anchovies, a three-dimensional geometric model was constructed, including the head, body, tail, dorsal fin, pelvic fin, vertebrae, and abdominal cavity. Using the foremost point of the fish head in the 3D geometric model as the origin O, the fish's central axis overlaps with the preset X-axis and the X-axis points towards the fish tail, the Y-axis is perpendicular to the X-axis and points to the left side of the fish body, and the Z-axis is perpendicular to the X-axis and points upwards, thus constructing a 3D geometric coordinate model.
6. The method for decomposition and cutting path planning in anchovy processing according to claim 1, characterized in that, Also includes: When cutting along the connection surface between the fish head and the fish body, the angle between the central axis of the fish head and the central axis of the fish body is obtained in real time. Sequentially adjust the angle between the fish head's central axis and the fish body's central axis, and the preset angle. Perform the difference calculation to obtain the first included angle difference; If the absolute value of the first included angle difference is greater than the preset first included angle threshold, then the first included angle difference is multiplied by the corresponding correction coefficient to obtain the first oblique angle correction value. Add the first bevel angle correction value to the bevel angle to obtain the corrected bevel angle; Continue cutting the fish head using the corrected angled cut.
7. The method for decomposing and cutting path planning during anchovy processing according to claim 1, characterized in that, Also includes: When cutting along the midline of the fish, the angle between the midline of the fish and the X-axis is obtained in real time; Sequentially adjust the angle between the fish's central axis and the X-axis to the preset angle. Perform the difference calculation to obtain the second included angle difference; If the absolute value of the second included angle difference is greater than the preset second included angle threshold, then the second included angle difference is multiplied by the corresponding correction coefficient to obtain the second oblique angle correction value. Add the second chamfer angle correction value to the chamfer angle to obtain the corrected chamfer angle; Continue cutting the fish body at the corrected angle.
8. The method for decomposition and cutting path planning in anchovy processing according to claim 1, characterized in that, Also includes: When cutting along the central axis of the fish, the distance from the current cutting position to the end of the central axis of the fish is obtained in real time; If the distance from the current cutting position to the end of the fish's central axis is less than or equal to a preset distance threshold, rotate the preset cutting angle. Adjust the cutting angle of the clamping blade according to the preset cutting angle to rotate and cut the fish until the internal organs and the fish body are separated.
9. The method for decomposition and cutting path planning in anchovy processing according to claim 1, characterized in that, The step of determining the connection surface between the fish head and the fish body specifically includes: Extract the following points from the three-dimensional geometric coordinate model of the anchovy: C1 (midpoint of the posterior margin of the skull), C2 (lowest point of the posterior margin of the left operculum), C3 (lowest point of the posterior margin of the right operculum), and C4 (lowest point on the dorsal side at the junction of the occipital condyle and the first vertebra). Using four feature points C1, C2, C3, and C4 as control points, a connection surface is generated by fitting a pre-defined triple bilinear B-spline surface. ,in These are parameters along the width of the fish's body. These are parameters along the height of the fish's body.