Method and System for Monitoring Sea Cucumber Quality during Sea Cucumber Processing

Abstract

The present invention relates to the field of visual detection technology, and particularly to a method and system for monitoring the quality of sea cucumbers during the processing of sea cucumbers. The present invention obtains a suspected papilla area of a sea cucumber image, analyzes the sharpness and position distribution of the suspected papilla area, and obtains the saliency of the suspected papilla area; sets and adjusts a basic scale, analyzes the distance relationship between suspected papilla areas based on the basic scale, obtains the uniformity of suspected papilla areas at different scales, corrects the saliency, and obtains the corrected saliency of suspected papilla areas at different scales; based on the corrected saliency at different scales, based on the CA saliency algorithm, obtains the final saliency of the suspected papilla area and the region of interest of the sea cucumber image; based on the final saliency, judges the quality of the region of interest, and obtains the quality of the sea cucumber. By means of the present invention, the determination accuracy of abnormal areas during the detection of the quality of sea cucumbers is improved, and damaged or poor-quality sea cucumbers can be accurately identified.

Description

Technical Field

[0001] The present invention relates to the technical field of visual detection, and particularly to a method and system for monitoring the quality of sea cucumbers during the processing of sea cucumbers. Background Art

[0002] During the processing of sea cucumbers, in order to ensure the nutritional components and market value of sea cucumbers, it is necessary to comprehensively detect the appearance of sea cucumbers. The papilla region of sea cucumbers often becomes an important basis for evaluating the quality of sea cucumbers. Especially when there is damage, the shape and distribution of the papilla will be affected, which may lead to nutrient loss and bacterial growth, thereby affecting the quality and shelf life of sea cucumbers. Therefore, during the processing, in order to ensure the quality of sea cucumbers, it is necessary to detect the integrity of the appearance of sea cucumbers in real time.

[0003] In the prior art, the CA saliency algorithm is usually used for sea cucumber image analysis to judge the integrity of the appearance of sea cucumbers. However, the traditional CA saliency algorithm has a low matching degree with sea cucumber images and cannot effectively judge the abnormal regions in sea cucumber images. Especially when there are changes in the spines, shape, and position of sea cucumbers, the existing CA saliency algorithm is difficult to accurately detect the damaged or abnormal regions of sea cucumbers. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a method and system for monitoring the quality of sea cucumbers during the processing of sea cucumbers.

[0005] According to the first aspect of the embodiments of the present invention, a method for monitoring the quality of sea cucumbers during the processing of sea cucumbers is provided, and the specific technical solution adopted is as follows:

[0006] Collect sea cucumber images;

[0007] Perform edge detection on the sea cucumber images to obtain suspected papilla regions;

[0008] Analyze the gray-scale features and shape features of the suspected papilla regions to obtain the sharpness of the suspected papilla regions;

[0009] Analyze the position distribution of the suspected papilla regions, and combine the sharpness to obtain the saliency of the suspected papilla regions;

[0010] Set and adjust the basic scale, and analyze the distance relationship between the suspected papilla regions based on the basic scale to obtain different uniformities of the suspected papilla regions corresponding to different scales;

[0011] According to the uniformity, correct the saliency to obtain different corrected saliencies of the suspected papilla regions corresponding to different scales;

[0012] Through iterative calculations using the CA significance algorithm based on different correction significances at different scales, the final significance of the suspected parapodium region and the region of interest of the sea cucumber image are obtained;

[0013] Based on the final significance, the quality of the region of interest is judged to obtain the quality of the sea cucumber.

[0014] In some embodiments of the present invention, analyzing the gray-scale features and shape features of the suspected parapodium region to obtain the sharpness of the suspected parapodium region includes:

[0015] Set the central axis of the sea cucumber image, and divide the suspected parapodium region into two cases: close to the central axis and far from the central axis;

[0016] When the suspected parapodium region is close to the central axis, analyze the gray-scale features of the suspected parapodium region to obtain the first sharpness of the suspected parapodium region;

[0017] When the suspected parapodium region is far from the central axis, analyze the shape features of the suspected parapodium region to obtain the second sharpness of the suspected parapodium region.

[0018] In some embodiments of the present invention, when the suspected parapodium region is close to the central axis, analyzing the gray-scale features of the suspected parapodium region to obtain the first sharpness of the suspected parapodium region includes:

[0019] When the suspected parapodium region is close to the central axis, construct the circumscribed rectangle of the suspected parapodium region, and obtain the length and width of the circumscribed rectangle;

[0020] Obtain the maximum value, mean value, and standard deviation of the gray-scale values of all pixel points in the suspected parapodium region, and combine the length and width of the circumscribed rectangle to obtain the first sharpness of the suspected parapodium region.

[0021] In some embodiments of the present invention, when the suspected parapodium region is far from the central axis, analyzing the shape features of the suspected parapodium region to obtain the second sharpness of the suspected parapodium region includes:

[0022] When the suspected parapodium region is far from the central axis, construct the circumscribed rectangle of the suspected parapodium region, and obtain the length and width of the circumscribed rectangle;

[0023] Obtain the pixel point in the suspected parapodium region that is farthest from the central axis as the reference point, connect the reference point with the two endpoints of the circumscribed rectangle that are close to the central axis respectively, and obtain the included angle between the two connecting lines;

[0024] According to the length and width of the circumscribed rectangle, and in combination with the cosine value of the included angle, obtain the second sharpness of the suspected parapodium region.

[0025] In some embodiments of the present invention, the position distribution of the suspected parapodium region is analyzed, and combined with the sharpness degree, the significance of the suspected parapodium region is obtained, including:

[0026] Obtain the pixel point with the farthest distance from the central axis in the suspected parapodium region as the reference point, and calculate the vertical distance between the reference point and the central axis;

[0027] Obtain the length of the short side of the sea cucumber image;

[0028] According to the vertical distance and the length of the short side, obtain the position distribution of the suspected parapodium region;

[0029] Weight the sharpness degree according to the position distribution to obtain the significance of the suspected parapodium region.

[0030] In some embodiments of the present invention, a basic scale is set and adjusted, and based on the basic scale, the distance relationship between the suspected parapodium regions is analyzed to obtain different uniformities of the suspected parapodium regions corresponding to different scales, including:

[0031] Set a basic scale, and based on the basic scale, construct a rectangular space in the sea cucumber image;

[0032] Analyze the distance relationship between the suspected parapodium regions included in the rectangular space to obtain the uniformity of the suspected parapodium regions corresponding to the basic scale;

[0033] Adjust the basic scale to obtain different uniformities of the suspected parapodium regions corresponding to different scales.

[0034] In some embodiments of the present invention, the basic scale includes a length basic scale and a width basic scale. Among them, the direction parallel to the central axis of the sea cucumber image is the direction of the length basic scale, and the direction perpendicular to the central axis of the sea cucumber image is the direction of the width basic scale.

[0035] In some embodiments of the present invention, based on the final significance, the quality of the concerned region is judged to obtain the sea cucumber quality, including:

[0036] Set a significance threshold;

[0037] Judge whether the final significance of the concerned region is greater than the significance threshold;

[0038] If so, mark the sea cucumber quality corresponding to the concerned region as poor.

[0039] According to the second aspect of the embodiments of the present invention, a sea cucumber quality monitoring system for the sea cucumber processing process is provided, including: a memory and a processor, wherein:

[0040] The memory is used to store program codes;

[0041] The processor is used to read the program codes stored in the memory and execute the method described in the first aspect of the embodiments of the present invention.

[0042] In some embodiments of the present invention, the processor includes:

[0043] A sea cucumber image acquisition and processing module, which is used to acquire sea cucumber images; and perform edge detection on the sea cucumber images to obtain suspected parapodium regions;

[0044] A sharpness analysis module, which is used to analyze the gray-scale features and shape features of the suspected parapodium regions to obtain the sharpness of the suspected parapodium regions;

[0045] A saliency analysis module, which is used to analyze the position distribution of the suspected parapodium regions, combine with the sharpness to obtain the saliency of the suspected parapodium regions; and set and adjust a basic scale, analyze the distance relationship between the suspected parapodium regions based on the basic scale to obtain different uniformities of the suspected parapodium regions corresponding to different scales; then, according to the uniformity, correct the saliency to obtain different corrected saliencies of the suspected parapodium regions corresponding to different scales; finally, based on the different corrected saliencies at different scales, perform iterative calculations through the CA saliency algorithm to obtain the final saliency of the suspected parapodium regions and the region of interest of the sea cucumber image;

[0046] A sea cucumber quality judgment module, which is used to judge the quality of the region of interest based on the final saliency to obtain the sea cucumber quality.

[0047] Compared with the prior art, the sea cucumber quality monitoring method and system provided by the present invention have the following beneficial effects:

[0048] By analyzing the sharpness and position distribution of the suspected parapodium area, the present invention obtains the significance of the suspected parapodium area, that is, through the precise analysis of the morphological characteristics of the suspected parapodium area in the sea cucumber image, combined with multi-dimensional information such as sharpness, reflection intensity, and position distribution. Compared with the traditional CA significance algorithm, the present invention can more accurately identify damaged or low-quality sea cucumbers. In addition, the present invention sets and adjusts the basic scale, analyzes the distance relationship between the suspected parapodium areas based on the basic scale, and obtains different uniformities of the suspected parapodium areas corresponding to different scales; then, according to the uniformity, the significance is corrected to obtain different corrected significances of the suspected parapodium areas corresponding to different scales; that is, by analyzing the distribution uniformity of the suspected parapodium area at different scales, the significance value is adjusted to reduce misjudgment caused by factors such as damage and missing, and improve the determination accuracy of abnormal areas during sea cucumber quality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0050] Figure 1 It is a schematic diagram of the basic process of a method for monitoring the quality of sea cucumbers during the sea cucumber processing provided by an embodiment of the present invention;

[0051] Figure 2 It is a schematic diagram of a sea cucumber area provided by an embodiment of the present invention;

[0052] Figure 3 It is a schematic diagram of the central axis of a sea cucumber area provided by an embodiment of the present invention;

[0053] Figure 4 It is a schematic diagram of the setting of the basic scale provided by an embodiment of the present invention;

[0054] Figure 5 It is a schematic diagram of the basic composition of a system for monitoring the quality of sea cucumbers during the sea cucumber processing provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0055] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manner, structure, features and effects of the method and system for monitoring the quality of sea cucumbers during the processing of sea cucumbers according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this invention belongs. Terms such as "including", "comprising" or any other variant thereof are intended to cover non-exclusive inclusion, such that a circuit structure, article or device including a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the presence of additional identical elements in the article or device including the element.

[0057] The specific scenario targeted by the embodiments of the present invention is as follows: During the processing of sea cucumbers, in order to ensure the nutritional components and market value of sea cucumbers, a comprehensive inspection of the appearance of sea cucumbers is required. The papillae area of sea cucumbers often becomes an important basis for evaluating the quality of sea cucumbers. Especially when there is damage, the shape and distribution of the papillae will be affected, which may lead to nutrient loss and bacterial growth, thereby affecting the quality and shelf life of sea cucumbers. Therefore, the object of the present invention is to provide a method and system for monitoring the quality of sea cucumbers during the processing of sea cucumbers, which can accurately monitor the appearance of sea cucumbers through image analysis technology. Especially in the case of damage or abnormality of sea cucumbers, it can efficiently detect and judge the quality of sea cucumbers. By performing morphological analysis on the papillae area in the sea cucumber image and combining features such as sharpness, reflection intensity, and uniformity of papillae arrangement, sea cucumbers with damage or poor quality can be accurately identified and screened, thereby improving the processing quality and storage stability of sea cucumbers.

[0058] The following specifically describes the specific solution of a method for monitoring the quality of sea cucumbers during the processing of sea cucumbers provided by the present invention in combination with the accompanying drawings.

[0059] Please refer to Figure 1 , which shows the basic process of a method for monitoring the quality of sea cucumbers during the processing of sea cucumbers provided by an embodiment of the present invention.

[0060] As Figure 1 shown, a method for monitoring the quality of sea cucumbers during the processing of sea cucumbers provided by an embodiment of the present invention specifically includes:

[0061] S100: Collect sea cucumber images.

[0062] Arrange lights and high-definition cameras at multiple angles above the conveyor belt used in sea cucumber processing to ensure that the lights can evenly illuminate every angle of the sea cucumber. For each sea cucumber, during the processing and transmission, collect multiple sea cucumber images through the high-definition camera, and ensure that the multiple images cover all surfaces of each sea cucumber.

[0063] For each acquired sea cucumber image, divide the sea cucumber area through semantic segmentation and perform grayscale processing on the sea cucumber image to finally obtain the grayscale image of each sea cucumber area, as Figure 2 shown.

[0064] S200: Perform edge detection on the sea cucumber image to obtain suspected parapodium areas.

[0065] Since the parapodium area and damaged area of the sea cucumber will cause differences from other areas in the sea cucumber image, the suspected parapodium (thorn) areas in each sea cucumber image can be obtained through edge detection. Specifically, since the thickness of the parapodium area on the sea cucumber is relatively low, in the sea cucumber image, the grayscale value of the parapodium area is higher than that of the main body area of the sea cucumber, and the parapodium area can be detected through edge detection; however, since the sea cucumber may be damaged resulting in the absence of the parapodium area, there is interference during edge detection. Therefore, for the sea cucumber image, obtain the edge area through edge detection, and record each obtained edge area as a suspected parapodium area.

[0066] S300: Analyze the grayscale features and shape features of the suspected parapodium areas to obtain the sharpness of the suspected parapodium areas.

[0067] In the existing CA saliency algorithm, by dividing the image into regions and calculating the differences between each divided region and other regions, the saliency value of each region is obtained. In the CA saliency algorithm, when the distance between adjacent regions is closer and the features are more similar, the saliency corresponding to this region is smaller; however, in the sea cucumber image, the positions of the parapodium areas are not adjacent, and there is a relatively large distance between two parapodium areas with non-parapodium areas in between; therefore, when processing the sea cucumber image through the existing CA saliency algorithm, the calculated saliency values are all relatively small, resulting in the existing CA saliency algorithm being unsuitable for processing sea cucumber images.

[0068] The papilla region in the sea cucumber image shows different states in the sea cucumber image when its position is different. The papillae of the sea cucumber are generally arranged in 4 rows or 6 rows, called 4-row spines or 6-row spines. Each row of spines (papillae) on the sea cucumber is roughly distributed on a line parallel to the central axis of the sea cucumber. The distance between each row of spines (papillae) of the sea cucumber and the central axis is different, and its state shown in the sea cucumber image is also different. The papilla region close to the central axis shows a top view in the sea cucumber image. Its image is mainly circular, and the tip of the papilla is located inside the circle and has a higher brightness. Therefore, for the suspected papilla region close to the central axis, the closer its edge approaches a circle and the higher the brightness of the reflective region of the tip of the papilla in the region, the more the suspected papilla region in this area belongs to the papilla region. The papilla region far from the central axis mostly shows a side view of the papilla, without a reflective region, but its side contour is more obvious. Therefore, for the suspected papilla region far from the central axis, the shape of the region formed by its edge is mostly slender and sharp, and its outermost end is an included angle opening inward. The larger the aspect ratio of this region and the narrower the included angle, the more the suspected papilla region belongs to the papilla region.

[0069] Based on the above analysis, in the embodiment of the present invention, first, by analyzing the gray-scale characteristics and shape characteristics of the suspected papilla region, the sharpness of the suspected papilla region is obtained. Further included are:

[0070] First, set the central axis of the sea cucumber image. The line passing through the center point of the sea cucumber region in each sea cucumber image along the long side direction can be used as the central axis of the sea cucumber region. Then, the suspected papilla region is divided into two cases: close to the central axis and far from the central axis.

[0071] Then, when the suspected papilla region is close to the central axis, analyze the gray-scale characteristics of the suspected papilla region to obtain the first sharpness of the suspected papilla region. The specific implementation method is as follows: when the suspected papilla region is close to the central axis, construct a circumscribed rectangle of the suspected papilla region (as Figure 3 shown), and obtain the length and width of the circumscribed rectangle; at the same time, obtain the gray-scale value of each pixel point in the suspected papilla region, then obtain the maximum value, average value and standard deviation of the gray-scale values of all pixel points in the suspected papilla region, and then combine the length and width of the circumscribed rectangle to obtain the first sharpness of the suspected papilla region. The formula for constructing the first sharpness of the suspected papilla region when it is close to the central axis is:

[0072]

[0073] In the formula, represents the first sharpness of the th suspected papilla region when it is close to the central axis; and respectively represent the The length and width of the circumscribed rectangle of a suspected parapodium area; Denote the maximum value of the grayscale values of all pixel points in the th suspected parapodium area; Denote the average value of the grayscale values of all pixel points in the th suspected parapodium area; Denote the standard deviation of the grayscale values of all pixel points in the

[0074] The aspect ratio of each suspected parapodium area is represented by . For each suspected parapodium area close to the central axis, the closer it is to the central axis, the greater the roundness, that is, the smaller its aspect ratio (closer to 1), indicating that the suspected parapodium area is more likely to belong to the parapodium area; Denote the reflectivity of each suspected parapodium area close to the central axis, Denote the ratio of the maximum brightness to the average brightness, reflecting the reflection intensity, reflecting the non-uniformity of the reflection. The reflectivity of the suspected parapodium area close to the central axis is obtained by combining the reflection intensity and the non-uniformity of the reflection. The greater the reflectivity of each suspected parapodium area close to the central axis, the more likely the suspected parapodium area close to the central axis belongs to the parapodium area.

[0075] Meanwhile, when the suspected parapodium area is far from the central axis, analyze the shape characteristics of the suspected parapodium area to obtain the second sharpness degree of the suspected parapodium area. The specific implementation method is as follows: when the suspected parapodium area is far from the central axis, construct the circumscribed rectangle of the suspected parapodium area (as shown in Figure 3 ), and obtain the length and width of the circumscribed rectangle; obtain the pixel point farthest from the central axis in the suspected parapodium area as the reference point, connect the reference point with the two endpoints close to the central axis in the circumscribed rectangle respectively to obtain the included angle between the two connecting lines; according to the length and width of the circumscribed rectangle, combined with the cosine value of the included angle, obtain the second sharpness degree of the suspected parapodium area when it is far from the central axis. The calculation formula for constructing the second sharpness degree of the suspected parapodium area when it is far from the central axis is:

[0076]

[0077] In the formula, Denote the sharpness degree of the th suspected parapodium area when it is far from the central axis; and respectively denote the length and width of the circumscribed rectangle of the th suspected parapodium area; Denote the included angle between the two connecting lines of the th suspected parapodium area; Denote the The cosine value of the included angle between two connecting lines of a suspected parapodium region.

[0078] By The aspect ratio representing each suspected parapodium region. For each suspected parapodium region far from the central axis, the farther it is from the central axis, the larger its aspect ratio, indicating that the sharpness of the suspected parapodium region far from the central axis is greater. Then it represents the cosine value of the included angle of each suspected parapodium region far from the central axis. When the included angle is smaller, the cosine value is larger, indicating that the sharpness of the suspected parapodium region far from the central axis is greater, and the greater the possibility that the suspected parapodium region far from the central axis belongs to the parapodium region.

[0079] S400: Analyze the position distribution of the suspected parapodium regions, and combine with the sharpness to obtain the significance of the suspected parapodium regions.

[0080] For the current scenario, since the CA significance algorithm cannot directly represent the significance of each suspected parapodium region based on the sea cucumber image, it is necessary to analyze the significance of each suspected parapodium region according to the position of each suspected parapodium region and its two sharpness levels.

[0081] Therefore, in the embodiments of the present invention, the position distribution of the suspected parapodium regions is analyzed, and combined with the sharpness, the significance of the suspected parapodium regions is obtained. The specific implementation method is as follows: Obtain the pixel point farthest from the central axis in the suspected parapodium region as the reference point, and calculate the vertical distance between the reference point and the central axis; obtain the short side length of the sea cucumber image, where the short side length is the short side length of the circumscribed rectangle of the sea cucumber region; according to the vertical distance and the short side length, obtain the position distribution of the suspected parapodium region; weight the sharpness according to the position distribution to obtain the significance of the suspected parapodium region. The significance calculation formula for constructing the suspected parapodium region is:

[0082]

[0083] In the formula, Represents the significance value of the th suspected parapodium region; Represents the vertical distance between the reference point and the central axis in the th suspected parapodium region; Represents the short side length of the width of this sea cucumber image; Represents the first sharpness level of the th suspected parapodium region when approaching the central axis; Represents the sharpness level of the th suspected parapodium region when far from the central axis; Represents The curve normalization function.

[0084] In this formula, by and the sharpness of each suspected parapodium region is weighted to calculate the significance value of each suspected parapodium region; the greater the sharpness of the suspected parapodium region, the more likely it is that the suspected parapodium region belongs to the parapodium region, and the smaller the significance should be. Therefore, when the distance between the suspected parapodium region and the central axis is farther, that is, the larger it is, the closer the suspected parapodium region is to the edge region, and the greater the impact of its sharpness on the significance. Vice versa.

[0085] S500: Set and adjust the basic scale, analyze the distance relationship between suspected parapodium regions based on the basic scale, and obtain the different uniformity of suspected parapodium regions corresponding to different scales.

[0086] In the original CA significance algorithm, after obtaining the significance value of each region, it is necessary to change the scale of each region to obtain the significance value of each region at different scales, and then obtain the region of interest in the image according to the CA significance algorithm; however, in the current scenario, the size scale of each suspected parapodium region in the sea cucumber image is fixed, so the original algorithm cannot be used for calculation.

[0087] In the current scenario, each parapodium region of the sea cucumber is roughly arranged in a pattern of 4 rows of spines and 6 rows of spines. Between each row of spines (parallel to the central axis), the distance distribution between the parapodium regions is relatively uniform (the distance parallel to the central axis, as Figure 3 shown). When there is a damaged area, it may cause the edge image to have a missing parapodium region in the sea cucumber image, resulting in an error in the uniformity of this row of spines. And although it is the same row of spines, the distances between them and the central axis in the direction perpendicular to the central axis are also not similar, and the parapodium regions are randomly distributed within a certain range. Therefore, it is necessary to set a length basic scale in the direction parallel to the central axis for the suspected parapodium region, and set a width basic scale in the direction perpendicular to the central axis for the suspected parapodium region. Then, through the distance distribution between the suspected parapodium regions in the rectangular region composed of the length basic scale and the width basic scale, calculate the difference between the distance between the suspected parapodium region in this rectangular region and its adjacent suspected parapodium region and the average distance of all suspected parapodium regions under this basic scale, as the uniformity of the suspected parapodium region under this basic scale, and the significance of the suspected parapodium region can be adjusted according to the uniformity.

[0088] Based on the above analysis, in the embodiments of the present invention, by setting and adjusting the basic scale, analyzing the distance relationship between suspected parapodium regions based on the basic scale, different uniformities of the suspected parapodium regions corresponding to different scales are obtained. The specific implementation method is as follows: Set the basic scale, where the basic scale includes a length basic scale and a width basic scale. Among them, the direction parallel to the central axis of the sea cucumber image is the direction of the length basic scale, and the specific length basic scale can be set as the length of the central axis within the sea cucumber region (as shown in Figure 4 ); the direction perpendicular to the central axis of the sea cucumber image is the direction of the width basic scale, and the specific width basic scale can be set as the maximum distance of a certain suspected parapodium region in this width direction (as shown in Figure 4 ); construct a rectangular space in the sea cucumber image based on the basic scale (length basic scale and width basic scale); analyze the distance relationship between the suspected parapodium regions included in the rectangular space (if half or more of the area of the suspected parapodium region is within the rectangular space, it is considered as the suspected parapodium region included in this rectangular space). Specifically, at this basic scale, calculate the mean and standard deviation of the distances between each suspected parapodium region in this rectangular space and its adjacent suspected parapodium regions. Among them, the distance between the suspected parapodium regions uses the projection position of the center point of each suspected parapodium region in the direction of the central axis as the standard point for calculating the distance, and the uniformity of the suspected parapodium region corresponding to the basic scale is obtained; adjust the basic scale, and the specific adjustment method can be, taking the basic scale as the standard, setting the change range to 1.1 or 1.2 times the basic scale, and continuously adjusting the size of the rectangular space to obtain different uniformities of the suspected parapodium regions corresponding to different scales. The formula for constructing the uniformity of the suspected parapodium region is:

[0089]

[0090] In the formula, represents the uniformity of the th suspected parapodium region under the th basic scale; and respectively represent the distances between the th suspected parapodium region under the th basic scale and its 2 adjacent suspected parapodium regions; represents the mean value of the distances between all suspected parapodium regions and their 2 adjacent suspected parapodium regions under the th basic scale; represents the standard deviation of the distances corresponding to all suspected parapodium regions under the th basic scale;

[0091] When is larger, it indicates that the distribution of this suspected parapodium region is more uneven, and it is more likely to be affected by damage or abnormality, and its significance needs to be adjusted.

[0092] S600: Modify the significance according to the uniformity to obtain different modified significances corresponding to different suspected parapodium regions at different scales.

[0093] According to the th basic scale and the th suspected parapodium region's uniformity, modify the significance to obtain the th basic scale and the th suspected parapodium region's modified significance. Furthermore, obtain multiple modified significances of all suspected parapodium regions at different scales. The formula for constructing the th basic scale and the th suspected parapodium region's modified significance is:

[0094]

[0095] In the formula, represents the modified significance value of the th suspected parapodium region; represents the significance value of the th suspected parapodium region; represents the uniformity of the th basic scale and the th suspected parapodium region.

[0096] The larger the

[0097] value is, the more uneven the distribution of the suspected parapodium region is, the more likely it is affected by damage or abnormality, and the more its significance needs to be adjusted.

[0098] S700: Based on the different modified significances at different scales, perform iterative calculations through the CA significance algorithm to obtain the final significance of the suspected parapodium region and the region of interest of the sea cucumber image.

[0099] After obtaining the different modified significances at different scales, input these multiple modified significances into the CA significance algorithm and perform iterative calculations through the CA significance algorithm to obtain the final significance of the suspected parapodium region and the region of interest (possible abnormal region) of the sea cucumber image.

[0100] S800: Based on the final significance, judge the quality of the region of interest to obtain the sea cucumber quality.

[0101] Based on the same inventive concept as the above method, this embodiment also provides a sea cucumber quality monitoring system for the sea cucumber processing process.

[0102] Please refer to Figure 5 , which shows the basic composition of a sea cucumber quality monitoring system provided by an embodiment of the present invention for the sea cucumber processing process.

[0103] As Figure 5 shown, a sea cucumber quality monitoring system for the sea cucumber processing process includes: a memory 10 and a processor 20, where:

[0104] The memory 10 is used to store program codes;

[0105] The processor 20 is used to read the program codes stored in the memory 10 and execute: collecting sea cucumber images; performing edge detection on the sea cucumber images to obtain suspected parapodium regions; analyzing the gray-scale features and shape features of the suspected parapodium regions to obtain the sharpness of the suspected parapodium regions; analyzing the position distribution of the suspected parapodium regions, and combining with the sharpness to obtain the saliency of the suspected parapodium regions; setting and adjusting a basic scale, analyzing the distance relationship between the suspected parapodium regions based on the basic scale to obtain different uniformities of the suspected parapodium regions corresponding to different scales; according to the uniformity, correcting the saliency to obtain different corrected saliencies of the suspected parapodium regions corresponding to different scales; based on the different corrected saliencies at different scales, performing iterative calculations through the CA saliency algorithm to obtain the final saliency of the suspected parapodium regions and the region of interest of the sea cucumber images; based on the final saliency, judging the quality of the region of interest to obtain the sea cucumber quality.

[0106] Further, the processor 20 includes a sea cucumber image acquisition and processing module 21, a sharpness analysis module 22, a saliency analysis module 23, and a sea cucumber quality judgment module 24. Where:

[0107] The sea cucumber image acquisition and processing module 21 is used to collect sea cucumber images; and perform edge detection on the sea cucumber images to obtain suspected parapodium regions;

[0108] The sharpness analysis module 22 is used to analyze the gray-scale features and shape features of the suspected parapodium regions to obtain the sharpness of the suspected parapodium regions;

[0109] The saliency analysis module 23 is configured to analyze the position distribution of the suspected parapodium regions, combine with the sharpness degree, and obtain the saliency of the suspected parapodium regions; and set and adjust the basic scale, analyze the distance relationship between the suspected parapodium regions based on the basic scale, and obtain the different uniformity of the suspected parapodium regions corresponding to different scales; then, according to the uniformity, correct the saliency to obtain the different corrected saliencies of the suspected parapodium regions corresponding to different scales; finally, based on the different corrected saliencies at different scales, perform iterative calculations through the CA saliency algorithm to obtain the final saliency of the suspected parapodium regions and the region of interest of the sea cucumber image;

[0110] The sea cucumber quality judgment module 24 is configured to judge the quality of the region of interest based on the final saliency and obtain the sea cucumber quality.

[0111] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0112] The various embodiments in this specification are all described in a progressive manner. The same or similar parts among the various embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. A method for monitoring the quality of sea cucumbers during the sea cucumber processing, characterized in that, The method includes: Collecting sea cucumber images; Performing edge detection on the sea cucumber images to obtain suspected parapodium regions; Analyzing the gray-scale features and shape features of the suspected parapodium regions to obtain the sharpness of the suspected parapodium regions; Analyzing the position distribution of the suspected parapodium regions and combining with the sharpness to obtain the saliency of the suspected parapodium regions; Setting and adjusting a basic scale, and analyzing the distance relationship between the suspected parapodium regions based on the basic scale to obtain different uniformities of the suspected parapodium regions corresponding to different scales; Correcting the saliency according to the uniformity to obtain different corrected saliencies of the suspected parapodium regions corresponding to different scales; Based on the different corrected saliencies at different scales, performing iterative calculation through the CA saliency algorithm to obtain the final saliency of the suspected parapodium regions and the region of interest of the sea cucumber images; Based on the final saliency, judging the quality of the region of interest to obtain the sea cucumber quality.

2. The sea cucumber quality monitoring method used in the sea cucumber processing process according to claim 1, characterized in that, Analyzing the gray-scale features and shape features of the suspected parapodium regions to obtain the sharpness of the suspected parapodium regions, including: Setting the central axis of the sea cucumber image, and dividing the suspected parapodium regions into two cases: close to the central axis and far from the central axis; When the suspected parapodium region is close to the central axis, analyzing the gray-scale features of the suspected parapodium region to obtain the first sharpness of the suspected parapodium region; When the suspected parapodium region is far from the central axis, analyzing the shape features of the suspected parapodium region to obtain the second sharpness of the suspected parapodium region.

3. The sea cucumber quality monitoring method during the sea cucumber processing process according to claim 2, characterized in that, When the suspected parapodium region is close to the central axis, analyzing the gray-scale features of the suspected parapodium region to obtain the first sharpness of the suspected parapodium region, including: When the suspected parapodium region is close to the central axis, constructing a circumscribed rectangle of the suspected parapodium region and obtaining the length and width of the circumscribed rectangle; Obtaining the maximum value, mean value and standard deviation of the gray-scale values of all pixel points in the suspected parapodium region, and combining with the length and width of the circumscribed rectangle to obtain the first sharpness of the suspected parapodium region.

4. The sea cucumber quality monitoring method during the sea cucumber processing process according to claim 2, characterized in that, When the suspected parapodium region is far from the central axis, analyzing the shape features of the suspected parapodium region to obtain the second sharpness of the suspected parapodium region, including: When the suspected parapodium region is far from the central axis, constructing a circumscribed rectangle of the suspected parapodium region and obtaining the length and width of the circumscribed rectangle; Obtaining the pixel point with the farthest distance from the central axis in the suspected parapodium region as a reference point, and connecting the reference point with the two endpoints close to the central axis in the circumscribed rectangle respectively to obtain the included angle between the two connecting lines; According to the length and width of the circumscribed rectangle and combining with the cosine value of the included angle, obtaining the second sharpness of the suspected parapodium region.

5. The method for monitoring the quality of sea cucumbers during the sea cucumber processing process according to claim 2, wherein, Analyzing the position distribution of the suspected parapodium regions and combining with the sharpness to obtain the saliency of the suspected parapodium regions, including: Obtaining the pixel point with the farthest distance from the central axis in the suspected parapodium region as a reference point, and calculating the perpendicular distance between the reference point and the central axis; Obtaining the length of the short side of the sea cucumber image; Based on the vertical distance and the short side length, obtain the position distribution of the suspected parapodium region; Based on the position distribution, weight the sharpness to obtain the significance of the suspected parapodium region.

6. The sea cucumber quality monitoring method during the sea cucumber processing process according to claim 1, wherein, Set and adjust the basic scale, and based on the basic scale, analyze the distance relationship between the suspected parapodium regions to obtain different uniformities of the suspected parapodium regions corresponding to different scales, including: Set the basic scale, and based on the basic scale, construct a rectangular space in the sea cucumber image; Analyze the distance relationship between the suspected parapodium regions included in the rectangular space to obtain the uniformity of the suspected parapodium regions corresponding to the basic scale; Adjust the basic scale to obtain different uniformities of the suspected parapodium regions corresponding to different scales.

7. The sea cucumber quality monitoring method during the sea cucumber processing process according to claim 6, characterized in that The basic scale includes a length basic scale and a width basic scale. Among them, the direction parallel to the central axis of the sea cucumber image is the direction of the length basic scale, and the direction perpendicular to the central axis of the sea cucumber image is the direction of the width basic scale.

8. The sea cucumber quality monitoring method during the sea cucumber processing process according to claim 1, characterized in that, Based on the final significance, judge the quality of the region of interest to obtain the sea cucumber quality, including: Set a significance threshold; Judge whether the final significance of the region of interest is greater than the significance threshold; If so, mark the sea cucumber quality corresponding to the region of interest as poor.

9. A sea cucumber quality monitoring system used in the sea cucumber processing process, characterized in that, The system includes: a memory and a processor, where: The memory is used to store program codes; The processor is used to read the program codes stored in the memory and execute the method according to any one of claims 1 to 8.

10. The sea cucumber quality monitoring system for sea cucumber processing according to claim 9, wherein The processor includes: A sea cucumber image acquisition and processing module, which is used to acquire a sea cucumber image; and perform edge detection on the sea cucumber image to obtain a suspected parapodium region; A sharpness analysis module, which is used to analyze the gray level characteristics and shape characteristics of the suspected parapodium region to obtain the sharpness of the suspected parapodium region; A significance analysis module, which is used to analyze the position distribution of the suspected parapodium region, combine the sharpness to obtain the significance of the suspected parapodium region; and set and adjust the basic scale, and based on the basic scale, analyze the distance relationship between the suspected parapodium regions to obtain different uniformities of the suspected parapodium regions corresponding to different scales; then according to the uniformity, correct the significance to obtain different corrected significances of the suspected parapodium regions corresponding to different scales; finally, based on the different corrected significances at different scales, perform iterative calculations through the CA significance algorithm to obtain the final significance of the suspected parapodium region and the region of interest of the sea cucumber image; A sea cucumber quality judgment module, which is used to judge the quality of the region of interest based on the final significance to obtain the sea cucumber quality.

Images