Crab quality identifying and grading device and method

Abstract

The invention discloses a crab quality identifying and grading device and method. The device comprises a conveying unit, an abdomen detecting unit, a back detecting unit, a weight detecting unit, a processing unit and a discharging unit. According to the method, firstly, through cooperation of double-belt differential of the conveying unit and a push rod, the standard detection posture of crabs with bellies facing downwards is automatically adjusted; then, sequentially collecting abdomen image information, and synchronously collecting back image information and weight information data; and the processing unit synthesizes the information, extracts gender and morphological characteristics, calculates fullness, and compares the gender and morphological characteristics with a preset standard rule base containing individual weight threshold values and sensory indexes so as to output judgment results of special-grade, first-grade, second-grade, third-grade or rejected products. The automatic and multi-dimensional comprehensive detection and accurate grading of the appearance, weight and key internal indexes of the crabs are realized, and the objectivity and efficiency of grading are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of automated testing technology for agricultural products, specifically to a crab quality identification and grading device and method. Background Technology

[0002] Traditional crab grading relies heavily on manual experience, judging grades by visually assessing size, weight, color, and vitality. This method suffers from high labor intensity, low efficiency, strong subjectivity, and inconsistent standards, making it difficult to meet the demands of large-scale, standardized commercial production. Existing automated grading equipment often focuses on mechanical screening based on single indicators such as weight or size, failing to comprehensively and non-destructively determine sensory characteristics that reflect core crab quality, such as carapace color, hair density, limb integrity, and internal indicators like plumpness. Furthermore, the unpredictable posture of crabs affects the accuracy of image acquisition; water stains and impurities in the production environment can contaminate detection elements, interfering with long-term operational stability. Therefore, there is an urgent need for an intelligent solution that can automatically, accurately, and efficiently perform multi-dimensional quality identification and grading, and adapt to complex working conditions. Summary of the Invention

[0003] The purpose of this invention is to provide a crab quality identification and grading device and method, which solves the technical problems in the prior art where crab grading relies on manual labor, is inefficient, has large subjective errors, and where automated equipment has limited functions and cannot make comprehensive and accurate judgments on multi-dimensional quality.

[0004] To achieve the above objectives, the present invention provides a crab quality identification and grading device, characterized in that it comprises: The conveying unit is used to transport crabs from the loading end to the unloading end; The abdominal detection unit includes a first camera for acquiring image information of the crab's abdomen; The back detection unit includes a second camera for capturing image information of the crab's back. The weight detection unit includes a weighing sensor located below the conveying unit, used to collect the weight information of each crab during transportation; The processing unit is configured to integrate the abdominal image information, the dorsal image information, and the weight information to grade the quality of a single crab. The feeding unit includes at least two feeding channels, allowing crabs of different grades to enter the corresponding feeding channels.

[0005] In a further technical solution, the first camera and the second camera are located on the upper and lower sides of the conveying unit, respectively.

[0006] In a further technical solution, at least two first cameras are provided, located on the upper and lower sides of the conveying unit respectively; at least two second cameras are provided, located on the upper and lower sides of the conveying unit respectively.

[0007] In a further technical solution, the conveying unit includes: The first conveying unit includes a first conveyor belt and a second conveyor belt arranged parallel to the crab conveying direction. The first conveyor belt and the second conveyor belt can be independently controlled in speed to adjust the crab conveying posture. The first conveyor belt and the second conveyor belt are provided with visual passage areas at intervals. The second conveying unit is connected to the end of the first conveying unit and is independently set on the weight detection unit, used to convey crabs from the first conveying unit to the unloading unit.

[0008] In a further technical solution, the abdominal detection unit also includes a first light source, the light emission direction of the first light source and the lens of the first camera are both directed toward the visual pathway area, and the first camera and the first light source are tunably mounted below the visual pathway area, and the mounting position can be adjusted along the conveying direction of the first conveying unit. A position-adjustable first sensor unit is installed on both sides of the first conveying section to detect the position of the crab, so as to control the abdomen detection unit to collect image information of the crab's abdomen. The bottom of the first sensor unit is fixedly connected to the abdominal detection unit so as to synchronously adjust the positions of the first sensor unit and the abdominal detection unit.

[0009] In a further technical solution, the back detection unit also includes a second light source, and the second light source and the second camera are tunably mounted on a bracket above the second conveying section; The second conveying unit includes a third conveyor belt, and two sides of the third conveyor belt are provided with position-adjustable second sensor units. The second sensor units are used to detect the position of the crab in order to control the weighing unit to weigh the crab and to control the back detection unit to collect image information of the crab's back.

[0010] In a further technical solution, the feeding unit is connected to the end of the second conveying section and also includes a swing-mounted diverter plate to divide the crabs into two paths; The feeding unit is equipped with a third sensor unit on both sides for controlling the swing of the flow divider plate.

[0011] A further technical solution also includes an interference cancellation unit, used to remove contaminants from the first camera, the second camera, and the second conveying unit, so as to ensure the accuracy of image acquisition and weight detection.

[0012] A method for identifying and grading crab quality, the method being based on a crab quality identification and grading device, includes the following steps: The system automatically adjusts the crab's transport posture so that its belly faces down. Obtain images of the crab's abdomen, back, and weight. By combining the abdominal image information, the back image information, and the weight information, and in conjunction with a pre-set standard rule library, the quality grade of each crab is determined. Based on the judgment result, a control command is sent to the diversion mechanism to control the crabs to enter the corresponding graded feeding channel; The pre-set standard rule base includes gender-differentiated individual weight class thresholds, sensory qualification conditions, rules for excluding deformities and injuries, and measurable index requirements.

[0013] A further technical solution involves combining the abdominal image information, the dorsal image information, and the weight information, along with a pre-set standard rule library, to determine the quality grade of each crab. Specifically: Based on abdominal image information, the sex, ventral color and gill area status of crabs are identified. Based on dorsal image information, the carapace color, smoothness, cheliped hair coverage, and leg integrity are identified. The length L of the cephalothorax is extracted from the back image information, and the fullness CF is calculated by combining it with the weight information Bw: CF = Bw / L 3 ; The standard rule base is invoked to mark individuals that are sensory unqualified, deformed, or damaged as to be removed; for qualified individuals, their quality level is determined by combining gender, weight, fullness, and boundary ambiguity compensation strategy, and the level includes special grade, first grade, second grade, or third grade.

[0014] In summary, the present invention has the following beneficial effects: it realizes full automation from posture adjustment, image acquisition, weighing, analysis to sorting and unloading, replacing the high-intensity and low-efficiency manual sorting, and makes judgments based on a unified and quantified standard rule base, completely eliminating subjective differences of human, ensuring the consistency and fairness of grading results within and between batches, and strongly supporting product standardization.

[0015] Breaking away from the limitations of traditional grading that relies solely on weight and rough appearance, this system integrates weight, size to calculate fullness, high-definition visual characteristics of the abdomen and back, gender, color, smoothness, down coverage, and integrity to construct a comprehensive quality assessment system. In particular, by using the fullness CF value, a key physiological indicator, the grading criteria have been deepened from "size" to "fullness," making the grading results more reasonable.

[0016] The system incorporates a dynamic attitude adjustment mechanism that automatically corrects feeding deviations, ensuring the reliability of the detection premises. For "boundary samples" at critical grade levels, an innovative fuzzy judgment strategy, such as fullness compensation, is introduced to make the grading decision more closely reflect complex realities.

[0017] The integrated interference cancellation unit can automatically clean key detection components, such as cameras, light sources, and weighing platforms, on a regular or as-needed basis. This effectively prevents the cumulative impact of environmental factors such as water stains, dirt, and salt stains on detection accuracy, reduces the frequency of manual maintenance, and ensures that the equipment maintains high reliability and stable detection performance even in humid and dusty aquatic product processing environments. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a first three-dimensional structural schematic diagram of the grading device of this application; Figure 2 This is a second three-dimensional structural schematic diagram of the grading device of this application; Figure 3 This is a third-dimensional structural schematic diagram of the grading device of this application; Figure 4 This is a top view of the grading device of this application without the housing; Figure 5 This is a schematic diagram of the fourth three-dimensional structure of the grading device of this application; Figure 6 This is a partial three-dimensional structural diagram of the interior of the box in this application; Figure 7 This is a schematic diagram of the side structure of the box inside this application; Figure 8 This is a schematic diagram illustrating the principle of the grading method in this application.

[0019] In the diagram: 100, base; 110, abdominal detection unit; 111, first camera; 120, back detection unit; 121, second camera; 130, weight detection unit; 131, weighing sensor; 140, processing unit; 150, unloading unit; 151, unloading channel; 152, flow divider; 200, conveying unit; 201, visual pathway area; 210, first conveying section; 211, first conveyor belt; 212, second conveyor belt; 213, first conveyor frame; 220, second conveying section; 221, third conveyor belt; 222, second conveyor frame; 223, second drive unit; 230, third conveying section; 231, third conveyor frame; 232, fourth conveyor belt; 233, third drive unit; 240, correction unit; 241, correction camera. ; 242, Electric push rod; 250, First sensor unit; 251, First signal transmitter; 252, First signal receiver; 260, Second sensor unit; 261, Second signal transmitter; 262, Second signal receiver; 270, Third sensor unit; 300, Housing; 301, Bracket; 302, Heat dissipation unit; 303, Door; 304, Emergency stop button; 400, Interference cancellation unit; 410, Cleaning mechanism; 411, Rodless cylinder; 412, Slider; 413, Lifting rod; 414, Roller brush; 415, Air inlet; 416, Air pump; 420, Blowing mechanism; 421, Slide plate; 422, First motor; 423, Threaded rod; 424, Second motor; 425, Electric telescopic rod; 426, Blowing head; 427, Blowing pump. Detailed Implementation

[0020] To more clearly illustrate the overall concept of this application, a detailed explanation is provided below with reference to the accompanying drawings.

[0021] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of this application is not limited to the specific embodiments disclosed below.

[0022] Furthermore, it should be understood in the description of this application that the terms "top", "bottom", "inner", "outer", "axial", "radial", "circumferential", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0023] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a communication connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0024] In this application, unless otherwise expressly specified and limited, the "above" or "below" of the second feature can mean that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. In the description of this specification, references to terms such as "an embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in one or more embodiments or examples.

[0025] like Figures 1-7 As shown, a crab quality identification and grading device is particularly suitable for crabs with hard shells, such as the Chinese mitten crab (Malva spp.), swimming crab, flower crab, and mud crab. The device includes: Conveying unit 200 is used to transport crabs from the loading end to the unloading end; The abdominal detection unit 110 includes a first camera 111 for acquiring image information of the crab's abdomen; The back detection unit 120 includes a second camera 121 for acquiring image information of the crab's back. The weight detection unit 130 includes a weighing sensor 131 located below the conveying unit 200, used to collect the weight information of each crab during transportation. Processing unit 140 is configured to perform quality grading on a single crab by integrating the abdominal image information, the back image information, and the weight information. The feeding unit 150 includes at least two feeding channels 151, allowing crabs of different grades to enter the corresponding feeding channels 151.

[0026] The grading device includes a base 100, an abdominal detection unit 110, a back detection unit 120, a weight detection unit 130, a processing unit 140, and a feeding unit 150, all of which are mounted on the base 100.

[0027] In one embodiment, the first camera 111 is located on the lower side of the conveying unit 200, and the second camera 121 is located on the upper side of the conveying unit 200. In this embodiment, both the first camera 111 and the second camera 121 can be visual cameras.

[0028] In another embodiment, the first camera 111 is located on the upper side of the conveying unit 200, and the second camera 121 is located on the lower side of the conveying unit 200. The first camera 111 is a visual camera, and the second camera 121 has a perspective imaging function, such as X-ray imaging technology. Low-dose X-rays penetrate the crab body. Different tissues of the crab (such as muscles, gonads, hepatopancreas, and shell) absorb X-rays to different degrees due to density differences, thereby forming an image with internal structural contrast on the detector. This allows for direct observation and quantification of the fullness and distribution of the gonads (roe / paste), the size of the hepatopancreas, the fullness of the meat, and the presence of lesions, parasites, or foreign objects inside. Alternatively, terahertz imaging technology can be used, utilizing terahertz waves of a specific band for imaging. Alternatively, both the first camera 111 and the second camera 121 can use visual cameras, but the material of the conveyor belt of the conveyor unit 200 can be modified to use a transparent material conveyor belt, such as a transparent plastic material conveyor belt, or an organic silicone with high optical transparency, so that image information can also be collected from crabs with their bellies facing upwards. This embodiment only provides a solution for detecting crabs when they are transported with their bellies facing upwards. In actual production, the crabs are mainly transported with their bellies facing downwards, and image information is obtained by taking pictures using the first camera 111 and the second camera 121 as visual cameras.

[0029] In another embodiment, two first cameras 111 are provided, located on the upper and lower sides of the conveying unit 200 respectively; two second cameras 121 are provided, located on the upper and lower sides of the conveying unit 200 respectively, which can detect crabs conveyed with their bellies facing down or up. In this embodiment, both first cameras 111 and both second cameras 121 can be vision cameras, and the conveyor belt should be made of transparent material. This embodiment can collect image information of crabs with their bellies facing up or down.

[0030] In one embodiment, the conveying unit 200 includes a first conveying section 210, a second conveying section 220, and a third conveying section 230.

[0031] The first conveying unit 210 includes a first conveyor belt 211 and a second conveyor belt 212 arranged parallel to the crab conveying direction. The first conveyor belt 211 and the second conveyor belt 212 can independently control their rotation speed to adjust the crab conveying posture. The first conveyor belt 211 and the second conveyor belt 212 are provided with visual passage areas 201 at intervals. The second conveying unit 220 is connected to the end of the first conveying unit 210 and is independently disposed on the weight detection unit 130. It is used to convey crabs from the first conveying unit 210 to the unloading unit 150.

[0032] Specifically, the first conveying unit 210 includes a first conveying frame 213, on which a first conveyor belt 211 and a second conveyor belt 212 are provided. A gap is provided between the first conveyor belt 211 and the second conveyor belt 212, which is a visual passage area 201. When the crab's abdomen is facing down, it is convenient for the abdomen detection unit 110 located on the lower side of the conveying unit 200 to take pictures.

[0033] The first conveyor belt 211 and the second conveyor belt 212 are respectively poweredly connected to two first drive units (not shown) mounted on the first conveyor frame 213. The first drive unit can be a direct drive motor or a servo motor. The two first drive units are poweredly connected to the rotating rollers in the first conveyor belt 211 and the second conveyor belt 212 respectively. The first drive unit can drive the first conveyor belt 211 and the second conveyor belt 212 to run synchronously, or it can drive the corresponding first conveyor belt 211 and the second conveyor belt 212 to run independently, so that the first conveyor belt 211 and the second conveyor belt 212 run at different speeds, thereby automatically adjusting the conveying posture of the crab.

[0034] Specifically, regarding the adjustment of the crab's posture, in one embodiment, the worker can manually place the crab on the first conveyor 210 in a preset posture. The preset posture can be as follows: when placing the crab, its head faces the crab conveying direction, its abdomen faces down, its back faces up, and its two legs are placed on the first conveyor belt 211 and the second conveyor belt 212 respectively. The line connecting the midpoint between the crab's two eyes and the midpoint of the crab's tail is located in the center of the visual pathway area 201. At this time, there is no need to adjust the crab's posture. The two first drive units are controlled to run synchronously, driving the first conveyor belt 211 and the second conveyor belt 212 to run synchronously, thereby ensuring that the crab is stably conveyed forward. In this embodiment, the abdomen detection unit 110 is located on the lower side of the visual pathway area 201. The abdomen detection unit 110 can collect the image information of the crab's abdomen through the visual pathway area 201. The back detection unit 120 is located above the conveyor unit and is used to detect the image information of the crab's back.

[0035] In another embodiment, a position correction unit 240 is provided on the first conveyor frame 213. The correction unit 240 includes two electric push rods 242 disposed on the first conveyor frame 213 and a position correction camera 241 located on the lower side of the first conveying section 210. The two electric push rods 242 are located on both sides of the visual passage area 201. When the crab's navel cannot be fully exposed in the visual passage area 201, for example, when the crab is not placed upright, specifically, when the crab is placed at an angle and the line connecting the midpoint between the crab's two eyes and the midpoint of the crab's tail is not parallel to the conveying direction, the crab's navel cannot be fully exposed in the visual passage area 201, and the correction camera 241 cannot fully expose the navel. The entire crab navel is detected. At this time, the image is captured by the position correction camera and then processed by the processing unit 140. The first conveyor belt 211 and the second conveyor belt 212 are controlled to rotate at different speeds. By using the speed difference between the two conveyor belts and the electric push rods on both sides, the crab is forced to automatically adjust its posture during the conveying process, thereby correcting the position of the crab. The line connecting the midpoint between the crab's two eyes and the midpoint of the crab's tail is parallel to the conveying direction, so that the crab navel is fully exposed in the visual passage area. The two rows of crab legs are located on the first conveyor belt 211 and the second conveyor belt 212 respectively, thus ensuring that the subsequent abdominal detection unit 110 can collect abdominal image information.

[0036] In one embodiment, a position-adjustable first sensor unit 250 is installed on both sides of the first conveying section 210 to detect the position of the crab so as to control the abdomen detection unit 110 to collect information about the crab's abdomen.

[0037] Specifically, a first sensor unit 250 is slidably mounted on the first conveyor frame 213. The first sensor unit 250 includes a first signal transmitter 251 and a first signal receiver 252. The first signal transmitter 251 and the first signal receiver 252 are located on both sides of the first conveyor frame 213, respectively, and are used to detect whether the crab has reached the position. The first signal transmitter 251 and the first signal receiver 252 are slidably disposed on both sides of the first conveyor frame 213. After the position is adjusted by sliding, they are tightened and fixed to the first conveyor frame 213 by bolts.

[0038] In one embodiment, the abdominal detection unit 110 further includes a first light source (not shown) for providing illumination. The light emission direction of the first light source and the lens of the first camera 111 are both directed toward the visual pathway region 201. The first camera 111 and the first light source are tunably mounted below the visual pathway region 201, and their mounting positions can be adjusted along the conveying direction of the first conveying section 210.

[0039] Specifically, the base 100 is provided with a plurality of bolt holes (not shown) arranged along the conveying direction. The first camera 111 can move along the conveying direction and is then fixed to the base 100 by bolt fastening.

[0040] In one embodiment, the bottom of the first sensor unit 250 is fixedly connected to the abdominal detection unit 110 to synchronously adjust the positions of the first sensor unit 250 and the abdominal detection unit 110.

[0041] Specifically, the first signal transmitter 251 and the first signal receiver 252 are fixed to the first camera 111 by connecting rods (not shown), so that the positions of the first signal transmitter 251, the first signal receiver 252 and the first camera 111 can be adjusted simultaneously.

[0042] In one embodiment, the back detection unit 120 further includes a second light source (not shown), and the second light source and the second camera 121 are tunably mounted on a bracket 301 above the second conveying section 220.

[0043] The second conveying unit 220 includes a third conveyor belt 221. The third conveyor belt 221 is provided with adjustable second sensor units 260 on both sides. The second sensor units 260 are used to detect the position of the crab so as to control the weight detection unit 130 to weigh the crab and control the back detection unit 120 to collect image information of the crab's back.

[0044] Specifically, a housing 300 is provided on the base 100, and a bracket 301 is fixed inside the housing 300. The processing unit 140 is fixed on the bracket 301. The second camera 121 and the second light source are slidably mounted on the bracket 301. After the position is adjusted, the second camera 121 and the second light source are fixed to the bracket 301 by bolts. A heat dissipation unit 302 is provided on the top of the housing 300 to dissipate heat from the various functional units inside the housing 300. A door 303 that can be opened is provided on one side of the housing 300, and an emergency stop button 304 is provided on the door 303.

[0045] The second conveying section 220 is connected to the end of the first conveying section 210 and includes a second conveying frame 222. The second conveying frame 222 is fixed on the weight detection unit 130. The weight detection unit 130 includes a weighing sensor 131, which can be a cantilever beam type weighing sensor. One end is fixed to the base 100, and the other end is used to support the second conveying frame 222. A third conveyor belt 221 is provided on the second conveying frame 222 and is driven by a second drive unit 223 fixed on the second conveying frame 222. The second sensor unit 260 includes a second signal transmitter 261 and a second signal receiver 262. The second signal transmitter 261 and the second signal receiver 262 are located on both sides of the second conveyor frame 222, respectively. The second signal transmitter 261 and the second signal receiver 262 are slidably mounted on the base 100. After adjusting the positions of the second signal transmitter 261 and the second signal receiver 262, the second signal transmitter 261 and the second signal receiver 262 are fixed to the base 100 by bolts.

[0046] In one embodiment, the third conveying section 230 is connected to the end of the second conveying section 220, and the unloading unit 150 is connected to the end of the third conveying section 230. The third conveying section 230 includes a third conveying frame 231 and a fourth conveyor belt 232, which is driven by a third drive unit 233 fixed on the third conveying frame 231.

[0047] The feeding unit 150 is connected to the end of the third conveying unit 230 and includes a swing-mounted diverter plate 152 for diverting the crabs into two paths. In this embodiment, the feeding channel 151 is provided with two paths, and the diverter plate 152 is located between the two feeding channels 151. The diverter plate 152 is driven by a diverter motor. By swinging the diverter plate 152, the third conveying unit 230 can transport the crabs to the corresponding feeding channel 151.

[0048] In another embodiment, the two feed channels 151 can be further split into two paths at the end, and then split by another diversion plate, so that the crabs can be divided into four levels and transported to four different feed channels respectively.

[0049] The feeding unit 150 is provided with a third sensor unit 270 on both sides for controlling the swing of the diverter plate. The third sensor unit 270 includes a third signal transmitter and a third signal receiver for detecting the position of the crab.

[0050] In one embodiment, the crab quality identification and grading device further includes an interference cancellation unit 400, which is used to remove contaminants from the first camera 111, the second camera 121 and the second conveying unit 220, so as to ensure the image acquisition accuracy of the first camera 111 and the second camera 121 and the detection accuracy of the weight detection unit.

[0051] Specifically, the interference cancellation unit 400 includes a cleaning mechanism 410 located above the second conveying part 220. The cleaning mechanism 410 is slidably mounted on the bracket 301 inside the housing 300. After adjusting the position of the cleaning mechanism 410, the cleaning mechanism 410 and the bracket 301 are fixed by bolts. The cleaning mechanism 410 includes a rodless cylinder 411 that can be slidably mounted. A slider 412 that can slide up and down is mounted on the rodless cylinder 411. A lifting rod 413 is fixed on the slider 412. The lifting rod 413 is perpendicular to the third conveyor belt 221. A roller brush 414 is rotatably mounted at one end of the lifting rod 413 near the third conveyor belt 221. Air inlets 415 are mounted on both sides of the roller brush. The air inlets 415 are connected to an air pump 416 fixed on the bracket 301 and are used to suck up the contaminants cleaned by the roller brush 414.

[0052] The interference cancellation unit 400 also includes a blowing mechanism 420 disposed below the first conveying section 210. The blowing mechanism 420 includes a sliding plate 421 that can slide along the crab conveying direction. A first motor 422 is fixed on the base 100. The first motor 422 is powered by a threaded rod 423, which is threadedly connected to the sliding plate 421. A second motor 424 is fixed on the sliding plate 421. One end of the second motor 424 is powered by an electric telescopic rod 425. The working principle of the electric telescopic rod 425 is the same as that of the electric push rod 242. It is also a type of electric push rod, capable of extending and retracting. For ease of distinction from the electric push rod, it is named electric telescopic rod 425. The electric telescopic rod 425 can extend upwards and rotate under the drive of the second motor 424. The top of the electric telescopic rod 425 can pass upwards through the visual passage area 201. A blower head 426 is fixed to the top of 425, and an air pump 427 is fixed on the base 100. The air pump 427 is connected to the blower head 426. The blower head 426 can swing up and down through a drive structure. The drive structure can be a swing motor, which drives the blower head 426 to swing up and down. The drive structure can also be an electromagnet structure. For example, the blower head 426 is installed in a frame through a pin. Electromagnets are provided on both the upper and lower sides of the frame. An iron structure is installed at the tail of the blower head. When the upper electromagnet is energized, it can attract the iron structure at the tail of the blower head 426, so that the blower head 426 is facing the abdominal detection unit 110. When the lower electromagnet is energized, the iron structure at the tail of the blower head 426 is attracted, so that the blower head 426 is facing the back detection unit 120. In this way, by controlling the operation of the air pump 427, the first camera 111 and the second camera 121 can be blown.

[0053] When it is necessary to blow the third conveyor belt 221, the control drive structure drives the blow head 426 to swing up and down, while the second motor 424 controls the electric telescopic rod 425 to rotate back and forth, so that the blow head blows the third conveyor belt 221, and the air inlet 415 draws in air.

[0054] When the device is working for a long time, dust may accumulate on the abdominal detection unit 110 or the back detection unit 120, affecting the accuracy of image information acquisition. The first camera 111 and the second camera 121 can be cleaned by controlling the direction of the blower head 426.

[0055] For example, the abdominal detection unit 110 is located below the first conveying section 210. By adjusting the position of the slide plate 421, and by adjusting the swing motor or controlling the corresponding electromagnet, the blow head 426 is directed toward the first camera 111, and the air pump 427 is controlled to work to perform the blowing operation. If the second camera 121 accumulates dust, the electric telescopic rod 425 is extended upward by controlling the swing motor or the corresponding electromagnet to work, controlling the blower head 426 to face the second camera 121, and then the air pump 427 works to blow and clean it.

[0056] If the third conveyor belt 221 needs cleaning, control the rodless cylinder 411 to work, causing the slider 412 to move down, thereby causing the lifting rod 413 to descend, and the roller brush 414 to stick to the third conveyor belt 221. Control the roller brush 414 to work, and at the same time control the air inlets 415 on both sides of the roller brush 414 to start sucking air, and control the electric telescopic rod 425 to rise, so that the blower head 426 is aligned with the third conveyor belt 221. At the same time, control the second motor 424 to work, thereby controlling the electric telescopic rod 425 to reciprocate, so that the blower head 426 cleans the third conveyor belt 221.

[0057] The processing unit 140 is electrically connected to the abdominal detection unit 110, the back detection unit 120, the weight detection unit 130, the correction unit 240, the feeding unit 150, the first conveying unit 210, the second conveying unit 220, the third conveying unit 230, and the interference cancellation unit 400, and the processing unit 140 is electrically connected to the human-machine interaction unit (not shown) installed on one side of the entire device.

[0058] The method of using this grading device is as follows: S1. System Preparation and Parameter Settings: Power on the device to initialize the processing unit 140, each drive unit, and the detection unit; The human-computer interaction unit allows users to select or input the expected specifications, target grading standards, and corresponding origin and seasonal parameter sets for the current batch of crabs through its human-computer interface.

[0059] As needed, the installation positions of the first sensor unit 250, the second sensor unit 260, the first camera 111 and the second camera 121 are slidably adjusted along the conveying direction to adapt to the size of the crab, thereby adapting to the optimal detection point, and the fixing bolts are tightened for mechanical locking. The interference cancellation unit 400 is activated to perform preliminary cleaning on the lenses of the first camera 111, the second camera 121, and the surface of the third conveyor belt 221.

[0060] Check and confirm that the first camera 111, the second camera 121, the weighing sensor 131, all light sources and all sensor units are working properly.

[0061] S2. Feeding and adjusting the posture of the crabs: The tied-up live crab is placed at the starting end of the first conveyor section 210 with its head facing the conveying direction. For posture adjustment, use any of the following methods: Method 1 (Manual Placement): The crabs are placed with their bellies facing down, with the line connecting the midpoint between their eyes and the midpoint of their tail parallel to the conveying direction, and this line is aligned with the center of the visual pathway area 201. The two rows of crab legs are placed on the first conveyor belt 211 and the second conveyor belt 212, respectively. Then, the two first drive units are controlled to operate synchronously to smoothly transport the crabs.

[0062] Method 2 (Automatic Correction): If the crab is not placed correctly, the position correction camera 241 detects that the crab's navel is not fully exposed in the visual pathway area 201; the processing unit 140 controls the first conveyor belt 211 and the second conveyor belt 212 to generate a speed difference based on the image analysis results, and coordinates with the electric push rods 242 on both sides to perform micro-push, using the speed difference and mechanical limit to force the crab to automatically adjust its position during the conveying process until its posture meets the detection requirements.

[0063] S3, Abdominal image acquisition: When the crab is transported to the visual pathway region 201, its body blocks the light beam emitted by the first sensor unit 250; The first sensor unit 250 sends a trigger signal to the processing unit 140; The processing unit 140 controls the first light source of the abdominal detection unit 110 to light up and triggers the first camera 111 to capture a high-definition image of the crab's abdomen.

[0064] S4. Back image acquisition and synchronous weighing: Crabs enter the second conveyor section 220 via the first conveyor section 210; When the crab fully enters the weighing area on the third conveyor belt 221, its body blocks the beam emitted by the second sensor unit 260. The second sensor unit 260 sends a trigger signal to the processing unit 140; Processing unit 140 synchronously performs the following operations: The weighing sensor 131 of the control weight detection unit 130 performs stable weighing to obtain the weight information of the crab. The second light source of the back detection unit 120 is turned on, and the second camera 121 is triggered to capture a high-definition image of the crab's back.

[0065] S5. Comprehensive Analysis and Hierarchical Decision-Making: Processing unit 140 receives and processes the abdominal image information, back image information and weight information obtained in steps S3 and S4; Processing unit 140 makes a comprehensive judgment based on a preset hierarchical algorithm and rule base, the judgment including: The sex of a crab can be identified based on information from its abdominal image.

[0066] The fullness is calculated based on the back image information and weight information, and sensory indicators such as carapace color, smoothness, and limb integrity are evaluated.

[0067] Weight, sex, fatness, and sensory scores are matched with preset thresholds for premium, first-class, second-class, and third-class.

[0068] For "boundary samples" that are at the critical level, a fuzzy judgment is made based on their plumpness or sensory score.

[0069] Output the final quality grade of the crab and the diversion command.

[0070] S6. Precise material distribution and feeding: The crab enters the third conveyor 230 via the second conveyor section 220 and moves toward the diversion plate 152; When the crab reaches the inlet of the feeding unit 150, the third sensor unit 270 detects its position. According to the grading result determined in step S5, the processing unit 140 controls the diverter plate 152 to swing to the corresponding angle; The crabs are conveyed by the third conveyor 230 and guided by the diverter 152 to fall accurately into the corresponding feeding channel 151, thus completing the grading.

[0071] S7. Periodic cleaning and maintenance: During continuous operation, the processing unit 140 automatically activates the interference cancellation unit 400 based on a preset cycle (e.g., after processing N crabs) or on the degree of cleanliness degradation determined by the camera images. The blower head 426 of the control blower mechanism 420 is moved to the side of the first camera 111, the second camera 121 or above the third conveyor belt 221 to perform blow cleaning.

[0072] The control cleaning mechanism 410 lowers the roller brush 414 to contact the surface of the third conveyor belt 221, and starts the roller brush rotation and the suction pump 416 to perform brushing and dust suction.

[0073] When the equipment is not in use for a long time or after the end of each workday, manually wipe and maintain all surfaces of the equipment, and check whether all fasteners are loose.

[0074] The conditions for activating the interference cancellation unit can also be based on predictive maintenance of the equipment's operating status: Real-time monitoring of current in each drive unit, zero-point drift of the weighing sensor, and illuminance attenuation data of the light source; When any monitoring data exceeds the normal threshold range, it is determined that the reliability of the relevant detection unit has decreased, and an equipment warning message is issued while performing cleaning and maintenance.

[0075] like Figure 8 The method for identifying and grading crab quality, as shown, includes: S100: Automatically adjusts the crab's transport posture so that the crab's belly faces down; S200: Obtain image information of the crab's abdomen, back, and weight. S300. Combining the abdominal image information, the back image information, and the weight information, and in conjunction with a pre-set standard rule library, determine the quality grade of each crab. S400: Based on the judgment result, send a control command to the diversion mechanism to control the crabs to enter the corresponding graded feeding channel; The pre-set standard rule base includes gender-differentiated individual weight class thresholds, sensory qualification conditions, rules for excluding deformities and injuries, and measurable index requirements.

[0076] Specifically, by combining the abdominal image information, the dorsal image information, and the weight information, and in conjunction with a pre-set standard rule base, the quality grade of each crab is determined, as follows: Based on abdominal image information, the sex, ventral color and gill area status of crabs are identified. Based on dorsal image information, the carapace color, smoothness, cheliped hair coverage, and leg integrity are identified. The length L of the cephalothorax is extracted from the back image information, and the fullness CF is calculated by combining it with the weight information Bw: CF = Bw / L 3 ; The standard rule base is invoked to mark individuals that are sensory unqualified, deformed, or damaged as to be removed; for qualified individuals, their quality level is determined by combining gender, weight, fullness, and boundary ambiguity compensation strategy, and the level includes special grade, first grade, second grade, or third grade.

[0077] Among them, S100 automatically adjusts the crab's transport posture so that the crab's belly faces down, specifically through automatic adjustment of the transport posture and standardized positioning: The first conveying section of the control conveying unit is started; Receive the initial posture image of the crab placed at the starting end of the first conveyor unit, which is acquired by the position correction unit; Based on the initial posture image, it is determined whether the crab is in a preset standard detection posture; the standard detection posture is: the abdomen is facing down, the line connecting the midpoint between the two eyes and the midpoint of the tail is parallel to the conveying direction, and the line corresponds to the center of the visual pathway area of ​​the first conveying part. If the determination is negative, a posture adjustment command is generated. The posture adjustment command is used to control the first conveyor belt and the second conveyor belt in the first conveying section to run at a different speed, and to coordinate with the control of the electric push rods set on both sides of the visual passage area, so that the crab automatically rotates and translates during the conveying process until it reaches the standard detection posture. If the determination is yes, or if the standard detection posture is achieved after adjustment, then the first conveyor belt and the second conveyor belt are controlled to run synchronously to smoothly transport the crab to the abdominal detection station.

[0078] Specifically, the acquisition of images of the crab's abdomen, back, and weight includes: S200, Synchronous / Sequential Triggered Acquisition of Multimodal Detection Information: When the first sensor unit detects that the crab has reached the preset abdominal detection point in the visual pathway area, the abdominal detection unit is triggered to collect image information of the crab's abdomen. The system controls the transport of the crab, after the abdominal image information has been acquired, to the second transport unit; When the second sensor unit detects that the crab has completely entered the weighing area on the second conveyor, the following operations are performed simultaneously: The weight detection unit is triggered to weigh the crab and obtain its weight information. The back detection unit is triggered to collect image information of the crab's back.

[0079] Specifically, by combining the abdominal image information, the dorsal image information, and the weight information, and in conjunction with a pre-set standard rule base, the quality grade of each crab is determined, as follows: S300, Comprehensive quality assessment based on rule and feature fusion: Receive and process the abdominal image information, dorsal image information and weight information obtained in step S200, and extract multi-dimensional features including sex, ventral color, carapace morphology, cheliped villi coverage, leg integrity, cephalothorax length and weight. The multi-dimensional features are input into a pre-set standard rule base for compliance verification; the standard rule base includes individual weight class thresholds differentiated by gender, sensory feature qualification standards, deformity and injury judgment rules, and fatness thresholds calculated based on the length and weight of the cephalothorax. Based on the compliance verification results and feature fusion analysis, the quality grade of each crab is output; the grade determination specifically includes: If it is identified as deformed or damaged, it is determined to be rejected; If the sensory characteristics are not up to standard, the product will be downgraded or rejected based on the non-compliance. For individuals that pass the sensory evaluation, a preliminary grade is made based on their sex, weight, and calculated body fatness. The grades include special grade, first grade, second grade, and third grade. For individuals whose weight falls within the critical range of adjacent weight classes, a fullness compensation mechanism is introduced: if their fullness is significantly better than the baseline value for the same weight class, they are allowed to be classified one level higher.

[0080] Based on the judgment result, a control command is sent to the diversion mechanism to control the crabs to enter the corresponding graded feeding channel, as follows: S400, Precise diversion control based on classification results: Based on the quality level output in step S300, generate the corresponding diversion control command; When the third sensor unit detects that a crab has arrived at the inlet of the feeding unit, the corresponding diversion plate is controlled to swing to the corresponding angle according to the diversion control command. After being transported, the crabs are guided by the diversion plate and accurately enter the feeding channel corresponding to the quality grade.

[0081] The "fatness threshold calculated based on carapace length and weight" mentioned in step S300 is derived from a specific product standard and is dynamically adjusted according to the fishing season; for crabs caught in a specific season, the fatness threshold serves as a mandatory basis for upgrading or downgrading.

[0082] Step S300 also includes a self-learning optimization sub-step for the hierarchical model: Collect sample data that were marked as “critical level” during the grading process or whose level was corrected after manual review, including their multi-dimensional features and the final confirmed level label; The collected sample data is used as an incremental training set to periodically fine-tune and optimize the algorithm model used for feature extraction or grade determination. The optimized model parameters are then updated and applied to subsequent grading processes to achieve continuous improvement in grading accuracy.

[0083] In one embodiment, the system's pre-defined crab standard rules are exemplified as follows:

[0084] When a worker places a male crab at the starting end of the first conveyor section, the first camera captures an image of its abdomen and identifies it as a male crab (with a narrow and elongated operculum), a grayish-white ventral surface, and no parasites in the gill area.

[0085] When the crab enters the second conveyor section, the second sensor unit is triggered. Weighing sensor reading: Bw = 285 g; The second camera captured images of the back: the carapace was dark green and smooth, the chelipeds had 82% hair coverage, and all eight legs were intact.

[0086] Overall quality assessment: Image information analysis extracted the length of the cephalothorax, L = 8.2 cm; Calculate fullness: CF = Bw / L 3 = 285 / 8.2 3 = 285 / 551.37 ≈ 0.517 g / cm³ 3 ; Rule base validation is called: Sex: Male crab; Weight: 285 g → Belongs to the first grade range (250–299 g); Fertility rating: 0.517 < 0.55 (Level 1 requirement) → Does not meet Level 1 fertility rating; However, it is in the critical range (only 15g away from the premium grade of 300g), and its sensory qualities are excellent. Enabling boundary fuzziness compensation strategy: Due to the velvet coverage rate >80% and the carapace ridge curvature being 12% higher than the average for the same weight, the grade is upgraded to Special Grade.

[0087] The processing unit generates a "special grade" diversion command. When the crab approaches the feeding port, the third sensor unit is triggered, and the diversion motor drives the diversion plate to swing, guiding the crab into the special grade channel.

[0088] Example of implementing a self-learning optimization mechanism: During the daily classification, the system marked 3 crabs as "critical samples": Sample A: Female crab, 218g, CF=0.49, originally classified as Grade 1, manually verified as Grade 1 (full of roe). Sample B: Male crab, 298g, CF=0.54, originally classified as Grade 1, artificially maintained as Grade 1; Sample C: Male crab, 245g, CF=0.53, originally classified as Grade II, artificially upgraded to Grade I (firm meat).

[0089] The system collects the feature vectors (weight, L, CF, villus coverage, abdominal fullness score, etc.) and correction labels of the above samples; Incremental training is started every night after shutdown: Fine-tuning was performed using a lightweight MLP classifier (learning rate 0.001, batch=3). The updated model assigns higher weight to the combination of "high fluff + high curvature".

[0090] The following day's test showed that the critical sample misclassification rate dropped from 8.7% to 3.2%.

[0091] Example of seasonal dynamic threshold application: Date: October 20, 2025 (within the best quality period); A female crab: weight 205g, CF=0.495, abdominal image fullness score 0.88; The system enables seasonal enhancement rules: Although CF < 0.51, it meets the criteria of "weight ≥ 200g + fullness ≥ 0.85"; It is directly classified as Grade A, without relying on actual measurement of gonadal index.

[0092] Through a self-learning mechanism, the accuracy of classification has steadily improved over long-term operation.

[0093] For any parts not mentioned in this application, existing technologies may be used or referenced.

[0094] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0095] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A crab quality identification and grading device, characterized in that, include: The conveying unit is used to transport crabs from the loading end to the unloading end; The abdominal detection unit includes a first camera for acquiring image information of the crab's abdomen; The back detection unit includes a second camera for capturing image information of the crab's back. The weight detection unit includes a weighing sensor located below the conveying unit, used to collect the weight information of each crab during transportation; The processing unit is configured to integrate the abdominal image information, the dorsal image information, and the weight information to grade the quality of a single crab. The feeding unit includes at least two feeding channels, allowing crabs of different grades to enter the corresponding feeding channels.

2. The crab quality identification and grading device according to claim 1, characterized in that, The first camera and the second camera are located on the upper and lower sides of the conveying unit, respectively.

3. The crab quality identification and grading device according to claim 1, characterized in that, The first camera is provided in at least two locations, which are respectively located on the upper and lower sides of the conveying unit; the second camera is provided in at least two locations, which are respectively located on the upper and lower sides of the conveying unit.

4. The crab quality identification and grading device according to claim 2, characterized in that, The conveying unit includes: The first conveying unit includes a first conveyor belt and a second conveyor belt arranged parallel to the crab conveying direction. The first conveyor belt and the second conveyor belt can be independently controlled in speed to adjust the crab conveying posture. The first conveyor belt and the second conveyor belt are provided with visual passage areas at intervals. The second conveying unit is connected to the end of the first conveying unit and is independently set on the weight detection unit, used to convey crabs from the first conveying unit to the unloading unit.

5. A crab quality identification and grading device according to claim 4, characterized in that, The abdominal detection unit also includes a first light source, the light emission direction of the first light source and the lens of the first camera are both directed toward the visual pathway area, and the first camera and the first light source are tunably mounted below the visual pathway area, and the mounting position can be adjusted along the conveying direction of the first conveying unit. A position-adjustable first sensor unit is installed on both sides of the first conveying section to detect the position of the crab, so as to control the abdomen detection unit to collect image information of the crab's abdomen. The bottom of the first sensor unit is fixedly connected to the abdominal detection unit so as to synchronously adjust the positions of the first sensor unit and the abdominal detection unit.

6. A crab quality identification and grading device according to claim 2, characterized in that, The back detection unit also includes a second light source, and the second light source and the second camera are adjustablely mounted on a bracket above the second conveying section; The second conveying unit includes a third conveyor belt, and two sides of the third conveyor belt are provided with position-adjustable second sensor units. The second sensor units are used to detect the position of the crab in order to control the weighing unit to weigh the crab and to control the back detection unit to collect image information of the crab's back.

7. The crab quality identification and grading device according to claim 1, characterized in that, The feeding unit is connected to the end of the second conveying section and also includes a swing-mounted diverter plate to divide the crabs into two streams; The feeding unit is equipped with a third sensor unit on both sides for controlling the swing of the flow divider plate.

8. The crab quality identification and grading device according to claim 1, characterized in that, It also includes an interference cancellation unit for removing contaminants from the first camera, the second camera, and the second conveying unit to ensure the accuracy of image acquisition and weight detection.

9. A method for identifying and grading the quality of crabs, the grading method being based on a crab quality identification and grading device according to any one of claims 1-8, characterized in that, The system automatically adjusts the crab's transport posture so that its belly faces down. Obtain images of the crab's abdomen, back, and weight. By combining the abdominal image information, the back image information, and the weight information, and in conjunction with a pre-set standard rule library, the quality grade of each crab is determined. Based on the judgment result, a control command is sent to the diversion mechanism to control the crabs to enter the corresponding graded feeding channel; The pre-set standard rule base includes gender-differentiated individual weight class thresholds, sensory qualification conditions, rules for excluding deformities and injuries, and measurable index requirements.

10. A method for identifying and grading crab quality according to claim 9, characterized in that: Based on the combined information from the abdominal image, the dorsal image, and the weight, and in conjunction with a pre-defined standard rule base, the quality grade of each crab is determined, specifically as follows: Based on abdominal image information, the sex, ventral color and gill area status of crabs are identified. Based on dorsal image information, the carapace color, smoothness, cheliped hair coverage, and leg integrity are identified. The length L of the cephalothorax is extracted from the back image information, and the fullness CF is calculated by combining it with the weight information Bw: CF = Bw / L 3 ; The standard rule base is invoked to mark individuals that are sensory unqualified, deformed, or damaged as to be removed; for qualified individuals, their quality level is determined by combining gender, weight, fullness, and boundary ambiguity compensation strategy, and the level includes special grade, first grade, second grade, or third grade.

Images