Prawn body length nondestructive measurement device and method based on computer vision

Abstract

The invention discloses a lossless prawn body length measurement device and method based on computer vision, and the method comprises the steps: obtaining an image of a prawn in a sliding process on a slide way, carrying out the preprocessing, and carrying out the instance segmentation of a prawn body in the obtained preprocessed image through a YOLOv11 instance segmentation model; and preliminarily determining head and tail points in combination with the curvature peak points after smoothing, judging whether candidate peak points meet preset conditions or not, if so, determining a final tail point, obtaining head and tail key points, and obtaining a growth curve by adopting a midline fitting method so as to determine the real body length of the prawn. According to the method, full-automatic and high-precision nondestructive measurement of the body length of the prawns is realized by combining an instance segmentation model and a geometric feature analysis technology, reliable data support is provided for dynamic monitoring and evaluation of growth in a culture process, and digital and precise development of aquaculture is assisted.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of aquaculture and computer vision technology, and particularly relates to a device and method for non-destructive measurement of the body length of prawns based on computer vision, which can measure the body length of Penaeus vannamei in real time and non-destructively. BACKGROUND

[0002] As one of the aquatic products with the largest scale of aquaculture and significant economic benefits worldwide, Penaeus vannamei has its body length as a key biological index for evaluating the growth state, health level and aquaculture benefits of the cultured objects. The body length data not only reflect the growth and development of the prawn body, but also serve as an important basis for regulating feed delivery, reasonably arranging the density of aquaculture and implementing precise management. The Food and Agriculture Organization of the United Nations recommends that Penaeus vannamei be measured for body length multiple times during the standard aquaculture period to ensure scientific decision-making and maximize benefits during the aquaculture process. However, the traditional measurement of the body length of prawns mainly relies on manual fishing, the use of a ruler or caliper for contact measurement, which not only consumes a large amount of human resources, but also causes human operation errors and stress reactions of prawns, resulting in damage to the prawn body, the risk of disease transmission and pollution of the aquaculture water environment. In particular, in a high-density aquaculture environment, manual measurement is inefficient and cannot meet the needs of modern intensive and intelligent aquaculture. Therefore, developing a real-time, non-destructive, automated and high-precision measurement technology for the body length of Penaeus vannamei has become an urgent technical innovation direction for the aquaculture industry.

[0003] The automatic body length measurement technology based on computer vision can facilitate rapid measurement and dynamic monitoring of the body length of prawns due to its non-contact, efficient and real-time characteristics. However, as a non-rigid aquatic organism, Penaeus vannamei naturally exhibits a highly flexible bending posture. Individuals often present complex morphologies with multiple angles and curvatures during the camera shooting process, especially when they are naturally swimming or affected by water flow. This non-linear three-dimensional bending projection onto a two-dimensional image poses a great challenge to traditional linear or key point-based methods for body length estimation. Simple straight-line measurement or key point connection often cannot accurately reflect the true length of the prawn body, resulting in measurement deviation. In addition, the bifurcated structure of the tail of Penaeus vannamei forms a complex geometric morphological feature. The tail fan is composed of multiple flexible appendages, and the outline area presents obvious sharp bifurcation and overlap, which makes it difficult for existing algorithms to maintain continuity and stability. The above two characteristics make the body length measurement of Penaeus vannamei face serious identification and fitting problems, limiting the widespread application of visual measurement technology in precise aquaculture. SUMMARY

[0004] The present application aims at the deficiencies of the prior art, and provides a computer vision-based nondestructive measurement device and method for shrimp body length.

[0005] The technical scheme adopted by the present application is as follows:

[0006] A computer vision-based nondestructive measurement device for shrimp body length comprises a chute, a camera, a variable frequency water pump and a flow rate adjusting module, a data processing unit and an auxiliary lighting device. The chute is in an inclined shape, and the upper and lower ends of the chute are used for allowing the shrimp to be measured to enter from the upper end, so that the shrimp bodies flow continuously in the direction of water flow in a non-stress state and flow out from the lower end. The camera is directly opposite the chute and is used to collect clear flowing shrimp images. The auxiliary lighting device is used to uniformly illuminate the measurement area to prevent environmental light changes from affecting image quality. The variable frequency water pump and the flow rate adjusting module are installed at the upper end inlet of the chute and are used to ensure that the shrimp passes through the measurement area in a single column at a suitable speed without being blocked, so as to ensure the stability and continuity of image acquisition. The data processing unit is connected with the camera and is used for real-time processing of received image data and body length measurement. Meanwhile, the data processing unit is connected with the variable frequency water pump and the flow rate adjusting module, and is used for cooperative control of the water pump and system alarm, so as to realize automatic measurement and efficient operation.

[0007] Based on the above measurement device, the present application provides a computer vision-based nondestructive measurement method for shrimp body length, which comprises the following steps:

[0008] Step 1: The data processing unit controls the variable frequency water pump and the flow rate adjusting module to be turned on, so that the shrimp to be measured is poured into the chute from the upper end inlet, and the camera captures the flowing video of the South American white shrimp in a non-stress state in real time, and transmits clear high-definition images to the data processing unit through a high-speed data interface.

[0009] Step 2: The data processing unit pre-processes the received original video, divides the video frame by frame, extracts the picture information of each frame, and uses Gaussian filtering to reduce environmental noise and enhance the edge contrast of the shrimp body.

[0010] Step 3: A YOLOv11 instance segmentation model is used to perform instance segmentation on the shrimp body in the pre-processed image, and a binary shrimp body mask is output.

[0011] Step 4: The contour point set {P0, P1, P2...P n} of the shrimp body is extracted from the mask, and a cubic B-spline curve is preferably used to smooth the contour point set. According to the recursively defined basis function, a smooth curve is constructed.

[0012]

[0013] Thus, the jagged edges and local irregularities in the YOLOv11 instance segmentation results are eliminated.

[0014] Step 5: Calculate the curvature values on the smoothed contour curve, identify the convex and concave features by the positive and negative changes of the curvature, classify the contour points according to the geometric morphological characteristics, and find the curvature peak points on the contour.

[0015] Step 6: Select the maximum value point from the curvature peaks as one of the key points at the head or tail (the first feature point), and then find the corresponding key point at the other end according to the contour arc length (i.e., select the point farthest from the first feature point on the contour as the second feature point among the remaining positive curvature peak points), to complete the preliminary positioning of the head and tail of the shrimp.

[0016] Step 7: For the complex bifurcated structure of the tail, adopt the adjacent peak merging strategy: for the feature point P i and its adjacent peak point P j , first calculate the arc length distance between the two points, if the distance between the two curvature peak points is less than the preset threshold value δ·D arc , then detect whether there is a negative curvature representing a concave in the contour region between the two points, and finally judge whether the curvatures of the two points are close, i.e. where δ is the distance threshold factor, k(P i ) is the curvature of P i , τ is the proportion factor, and D arc is the arc length of the whole contour. If all three conditions are met, it is considered that the two are adjacent peak point pairs. Subsequently, the median point between the two points is used to replace the original two bifurcated sharp points, realizing the unified representation of the shrimp tail bifurcated structure and reducing the influence of the measurement accuracy caused by the bifurcated structure. After obtaining the finally determined tail point, the other point far from the tail point in step 6 is the finally determined head point;

[0017] Step 8: With the head and tail key points as boundaries, use the midline fitting method to generate a continuous and smooth body length curve. Specifically: from the head point to the tail point, divide the complete contour into two sub-contours according to the clockwise and counterclockwise paths, calculate the arc lengths of the two parts of the contour and sample them equally, obtain the corresponding sampling points, connect each pair of sampling points and calculate the midpoint of their geometric line, and finally connect all the midpoints in order with a smooth curve, which can obtain a complete and continuous contour midline. After calculating the pixel length of the body length curve, multiply it by the proportion coefficient to obtain the real body length of the shrimp, wherein the proportion coefficient is the built-in parameter of the camera.

[0018] Step 9: The data processing unit adjusts the flow rate of the frequency conversion pump through the control interface according to the measurement results to ensure measurement stability. The measurement data is uploaded to the display end and the aquaculture management system in real time, supporting historical data storage and analysis.

[0019] The device of the present application adopts a slide, a camera, a frequency conversion water pump, a flow rate adjustment module, a data processing unit, and an auxiliary lighting device to form a complete automatic measurement device, which can measure the body length of prawns in various postures and specifications in real time and non-destructively.

[0020] The beneficial effects of the present application are:

[0021] The non-destructive measurement device for the body length of Penaeus vannamei has a simple structure and is easy to operate, and is suitable for any breeding environment. The advanced visual recognition and curvature analysis technology is used to realize the rapid and accurate measurement of the body length of prawns. The system can adapt to the natural bending posture of prawns and the tail fork structure, significantly improving the accuracy and stability of measurement. The non-contact measurement method effectively avoids the damage and stress reaction of traditional manual measurement of prawn body, ensuring the safety and hygiene of the breeding environment. At the same time, the device supports real-time high-throughput detection, greatly improving the efficiency of breeding management, and is suitable for large-scale and intelligent Penaeus vannamei breeding production. The technology not only improves the automation and intelligence level of measurement, but also provides reliable data support for breeders, and has good popularization prospect and application value. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a structure diagram of a non-destructive measurement device for the body length of prawns based on computer vision in an embodiment of the present application.

[0023] Figure 2 is a schematic diagram of instance segmentation of prawn body in an embodiment of the present application.

[0024] Figure 3 is a result diagram of curvature calculation. Figure 2 is a result diagram of curvature calculation.

[0025] Figure 4a is a result diagram of curvature calculation.

[0026] Figure 4b is a result diagram of curvature calculation.

[0027] Figure 5a is a result diagram of curvature calculation.

[0028] Figure 5b is a result diagram of curvature calculation.

[0029] Figure 6aThis is a schematic diagram of the forked structure treatment of the shrimp tail in an embodiment of the present invention.

[0030] Figure 6b In this embodiment of the invention, the median point of the arc length is used as the final determined tail point.

[0031] Figure 7a This is a schematic diagram of two contour lines formed by using the first and last key points as boundaries in an embodiment of the present invention.

[0032] Figure 7b This is a schematic diagram of equal arc length sampling of two contour parts in an embodiment of the present invention.

[0033] Figure 7c This is a schematic diagram of taking the midpoint after connecting corresponding sampling points on two contours in an embodiment of the present invention.

[0034] Figure 7d This is the body length curve obtained in the embodiments of the present invention. Detailed Implementation

[0035] The following description, in conjunction with the accompanying drawings, provides a more detailed account of the computer vision-based non-destructive measurement device and method for the body length of Litopenaeus vannamei.

[0036] According to a specific embodiment of the present invention, a computer vision-based non-destructive measurement device for shrimp body length is as follows: Figure 1 As shown, the overall structure of this automatic shrimp measuring device is a closed trapezoidal box. The top of the box features a large and deep open concave pool (shrimp inlet pool). The bottom of the concave pool transitions downwards to a transparent and smooth slide 1, forming a sloping measurement area that guides the shrimp smoothly to the measurement area for body length measurement. A high-definition camera 2 is fixed above the slope, capable of capturing images of the slope plane. Multiple LED strip lights 3 are evenly distributed along the slope surface, conforming to the lower side of the slope to ensure uniform illumination. A variable frequency water pump and flow rate adjustment module 4 are installed at the connection between the shrimp inlet of the concave pool and the upstream of the slope, allowing adjustment of the shrimp's entry speed and interval via a mechanical gate or vibration device. The end of the slope gradually transitions to a horizontal outlet slide, with the horizontal section extending outside the box to ensure the shrimp smoothly enter the external tank or water body. A data processing and control system 5 is installed directly below the sloping measurement area inside the device, and is sealed to the camera, lighting, and flow rate adjustment module to complete the automatic measurement operation.

[0037] The above-mentioned device was used to perform non-destructive measurement of the body length of Litopenaeus vannamei. The measurement method included the following steps:

[0038] Step 1: The industrial camera captures real-time dynamic images of shrimp flowing through the measuring area of ​​the slide. The image format is RGB color, with a resolution of 1920×1080, and is transmitted to the data processing unit.

[0039] Step 2: The data processing unit pre-processes the received raw video, extracts the picture information of each frame, and uses Gaussian filtering to reduce environmental noise and enhance the edge contrast of the shrimp body. The standard Gaussian kernel function is:

[0040]

[0041] where σ = 1.0 is the standard deviation, i, j are two input vectors, and the filter window size is 5x5.

[0042] Step 3: Call the trained YOLOv11 deep learning model to perform instance segmentation on the pre-processed image. The model inference output contains a binary mask for each shrimp body. The YOLOv11 instance segmentation model includes a C2PSA component that introduces a spatial attention mechanism to accurately identify the outline of the shrimp in complex environments, even in partially occluded situations. Second, the model significantly reduces computational overhead through the optimized design of the C3k2 block, with a segmentation speed of 82.0 FPS, allowing near real-time processing on medium configuration hardware and meeting the real-time requirements of high-throughput detection. Finally, YOLOv11 provides multiple scale models with a parameter size of only 22.3M, which has a clear lightweight advantage compared to other models and is more suitable for deployment in resource-constrained aquaculture monitoring environments. The experimental test results of each model are as follows, and YOLOv11 is the best choice in the application scenario of the present application.

[0043] Model mAP = 0.5:0.95 Params / M FPS Mask R-CNN 0.751 44.0 17.2 YOLACT 0.649 34.7 24.4 SOLOv2 0.753 46.2 15.4 Mask2Former 0.776 44.0 4.14 QueryInst 0.762 172.45 4.55 CondInst 0.765 33.98 11.8 YOLOv8 0.742 22.7 80.9 YOLOv11 0.741 22.3 82.0

[0044] As shown in Figure 2 , the output mask is as follows:

[0045]

[0046] Step 4: Extract the shrimp body contour point set {P0, P1, P2...P n} from the mask. The contour is a closed curve. There are sawteeth and local noise in the contour, and a cubic B-spline curve is used to smooth the contour point set. Its inherent local control point feature can accurately fit the head, abdomen, and tail of the shrimp, etc. without affecting the overall contour. Second, the C 2 continuity of the cubic B-spline curve combines the biological morphological characteristics of the shrimp, which can generate a natural and smooth curve, effectively eliminating the jagged edges in the segmentation results, while maintaining the reasonable bending characteristics of biology. Third, in the real-time processing scenario of large quantities of images, the recursive calculation method of the B-spline curve can significantly reduce the computational overhead, meeting the rapid processing requirements, which is difficult to achieve by other smoothing methods. The parameterized form is expressed as:

[0047]

[0048] where u ∈ [0, 1] is a normalization parameter, N i,p(u) is the p-th B-spline basis function. For high order basis functions, N i,p(u) can be defined recursively:

[0049]

[0050] N i,p (u) = [(u - t i ) / (t i+p -t i )] · N i,p-1 (u) + [(t i+p+1 -u) / (t i+p+1 -t i+1 )]

[0051] · N i+1,p-1 (u)

[0052] where t i is the knot value in the knot vector. The weights and the knot vector are chosen uniformly to guarantee the C 2 continuity of the curve, which eliminates the jaggedness and effectively preserves the morphological characteristics of the shrimp body. The smoothed effect is shown in Figure 3 .

[0053] Step 5: Calculate the curvature value on the smoothed contour curve:

[0054] x'(u) ≈ [x(u + Δu) - x(u - Δu)] / (2Δu)

[0055] x"(u) ≈ [x(u + Δu) - 2x(u) + x(u - Δu)] / (Δu) 2

[0056]

[0057] where k is the curvature, x' and y' are the first-order derivatives of x and y, respectively, and x" and y" are the second-order derivatives. The positive and negative changes of curvature are used to identify convex and concave features, and the curvature peak points on the contour are found. Positive curvature corresponds to convex features, which are often the head or tail tip of the shrimp, and negative curvature corresponds to concave features, which are often the abdomen of the shrimp in a curved posture, so as to realize the positioning of the head and tail points in the subsequent process, as shown in Figure 4a and Figure 4b .

[0058] Step 6: Select the maximum point from the curvature peaks as one of the key points at the head or tail, which corresponds to the most prominent convex region, and then select the point farthest from the first feature point on the contour from the remaining positive curvature peaks as the second feature point. This operation is based on the biological characteristics of the Penaeus vannamei, in which the arc length between the head and tail points accounts for about half of the total length of the contour. The two feature points thus obtained are the head and tail points of the shrimp, as shown in FIG. 6. Figure 5a , Figure 5b .

[0059] Step 7: For the complex bifurcated structure of the tail, the adjacent peak merging strategy is adopted: for the feature point P i and its adjacent peak point P j , first calculate the arc length distance between the two points, if the distance between the two curvature peak points is less than the preset threshold value δ·D arc , then detect whether there is a negative curvature in the contour region between the two points, if there is, finally judge whether the curvatures of the two points are close, i.e. where δ is the distance threshold factor, k(P i ) is the curvature of P i , τ is the proportion factor, and D arc is the arc length of the whole contour. If all three conditions are met, it is considered that the two are adjacent peak point pairs. Subsequently, the arc length median point between the two points is used to replace the original two bifurcated sharp points, realizing the unified representation of the shrimp tail bifurcated structure and reducing the influence of the bifurcated structure on the measurement accuracy. Thus, the arc length median point is determined as the final tail point, and the other point far from the point in step 6 is the final head point. As shown in FIG. 7 and FIG. 8. Figure 6a and Figure 6b .

[0060] Step 8: With the head and tail key points as boundaries, the complete contour is divided into two segments of sub-contours from the head point to the tail point according to the clockwise and counterclockwise paths, respectively corresponding to the "upper side" (or left side) and "lower side" (or right side) curves of the shrimp body. The arc lengths of the two parts of the contour are calculated and equi-arc length sampling is performed, respectively, to obtain the corresponding pairs of sampling points. Since the left and right side (or upper and lower) corresponding sampling points are approximately at the same segment position biologically, when connecting each pair of sampling points and calculating the midpoint of their geometric connecting line, these midpoints will naturally fall on the center position of the shrimp body. Finally, all the midpoints are connected by a smooth curve in order to obtain a complete and continuous contour centerline. After calculating the pixel length of the body length curve, the real body length of the shrimp is obtained by multiplying the proportion coefficient. As follows:

[0061]

[0062] L real = L pixel × S scale

[0063] wherein the proportionality factor S scale is an intrinsic parameter of the camera. Figure 7a-7d As shown.

[0064] Step 9: The data processing unit adjusts the flow rate of the variable frequency pump through the control interface according to the measurement results to ensure measurement stability. The measurement data is uploaded to the display end and the aquaculture management system in real time, supporting historical data storage and analysis.

[0065] The above disclosure is only a specific embodiment of the present application, but the present application is not limited thereto. Any modification made by those skilled in the art without departing from the present application shall be considered as falling within the scope of the present application.

Claims

1. A computer vision based non-invasive shrimp body length measuring device, characterized in that, The device comprises a chute, a camera, a variable frequency water pump and a flow rate adjusting module, a data processing unit and an auxiliary lighting device; the chute is in an inclined shape, the upper end is an inlet and the lower end is an outlet, the prawns to be measured enter from the upper end with the water body, the prawn bodies continuously flow in the direction of the water flow in a non-stress state and flow out from the lower end, the camera is opposite to the chute to collect clear flowing prawn images, the auxiliary lighting device is used to uniformly illuminate the measurement area to prevent environmental light changes from affecting image quality, the variable frequency water pump and the flow rate adjusting module are installed at the inlet of the upper end of the chute to ensure that the prawns pass through the measurement area without being blocked by adjusting the water flow rate, the data processing unit is connected with the camera and is used for real-time processing of received image data and body length measurement, and the data processing unit is connected with the variable frequency water pump and the flow rate adjusting module to cooperatively control the water pump and system alarm.

2. A non-destructive method for measuring the body length of shrimp based on computer vision, characterized in that, The method comprises the following steps: obtaining images of prawns sliding on the chute, performing pretreatment, performing instance segmentation on the prawn bodies in the obtained pretreated images by using a YOLOv11 instance segmentation model, performing smoothing processing, combining curvature peak points to preliminarily determine head and tail points, judging whether the candidate peak points meet predetermined conditions, determining the final tail point if the conditions are met, obtaining the finally determined head and tail key points, obtaining a growth curve by using a median fitting method, and determining the real body length of the prawn.

3. The non-destructive method for measuring body length of shrimp based on computer vision according to claim 2, wherein, The method comprises the following steps based on the device of claim 1: Step 1: the data processing unit controls the variable frequency water pump and the flow rate adjusting module to be turned on, the prawns are put into the inlet at the upper end of the chute, the camera captures the flowing video of the prawns in a non-stress state in real time, and the video is transmitted to the data processing unit; Step 2: the data processing unit performs pretreatment on the received original video, divides the video frame by frame to extract the picture information of each frame, and uses Gaussian filtering to reduce environmental noise and enhance the contrast of the edges of the prawn bodies; Step 3: the YOLOv11 instance segmentation model is used to perform instance segmentation on the prawn bodies in the pretreated images obtained in step 2, and a binary prawn mask is output; Step 4: Extract the shrimp body contour point set {P0, P1, P2...P n} from the mask, and smooth it using a b-spline curve, thereby eliminating the jagged edges and local irregularities in the YOLOv11 instance segmentation result; Step 5: the curvature values are calculated on the smoothed contour curve, the convex and concave features are recognized by using the positive and negative changes of the curvature, so that the contour points are classified according to the geometric morphological characteristics, and then the positive curvature peak points on the contour are obtained by screening; Step 6: the peak point with the maximum curvature is identified as the first feature point from the peak points obtained in step 5, then the point farthest from the first feature point on the contour is selected as the second feature point from the remaining positive curvature peak points, and the preliminarily determined head and tail points are obtained; Step 7: the double-peak feature of the contour curvature curve caused by the complex bifurcated structure of the tail of the prawn, when the two candidate peak points meet the predetermined conditions, the arc length median point is used to replace the two bifurcated sharp points of the original tail to realize the unified representation of the prawn tail bifurcated structure, and the finally determined tail point is obtained, and the other point far from the tail point in step 6 is the finally determined head point; Step 8: based on the head and tail key points identified in steps 6 and 7, a continuous and smooth body length curve is generated by using a median fitting method, the pixel length of the body length curve is calculated, and the real body length of the prawn is obtained by multiplying the pixel length by a proportionality coefficient.

4. The non-destructive method for measuring body length of shrimp based on computer vision according to claim 3, characterized in that, In step 4, the contour point set is smoothed by using three B-spline curves, and a smoothing curve is constructed according to the base function defined recursively. where n is the number of profile points, P i is the i-th profile point, N i,p (u) is the p-th B-spline basis function.

5. The non-destructive method for measuring body length of shrimp based on computer vision according to claim 3, wherein, In step 5, curvature values are calculated on the smoothed contour curve, and the concave-convex features are determined according to the positive and negative curvature values, wherein the positive curvature corresponds to the convex feature, and the negative curvature corresponds to the concave feature, which are used for subsequent head-tail point positioning.

6. The non-destructive method for measuring body length of shrimp based on computer vision according to claim 3, wherein, In step 6, the head-tail point positioning method is as follows: a peak point with the maximum curvature is identified as a first feature point from the peak points obtained in step 5, then a point with the longest distance from the first feature point on the contour is selected as a second feature point from the remaining positive curvature peak points, and thus the two feature points obtained are the preliminarily determined head-tail points of the prawn.

7. The non-destructive method for measuring body length of shrimp based on computer vision according to claim 3, wherein, The merging strategy of adjacent peaks in step 7 is as follows: for feature point P i and its adjacent peak point P j , first, the arc length distance between the two points is calculated, if the distance between the two curvature peak points is less than the preset threshold value δ·D arc , then the contour region between the two points is detected to see if there is a negative curvature representing a concave, if there is, finally, the curvatures of the two points are judged to see if they are close, that is where δ is the distance threshold factor, k(P i ) is the curvature of P i , τ is the proportional factor, and D arc is the arc length of the whole contour; if all three conditions are met, it is considered that the two are adjacent peak point pairs; subsequently, the median point of the arc length between the two points is used to replace the original two bifurcation sharp points, the unified representation of the shrimp tail bifurcation structure is realized, the influence of the measurement accuracy caused by the bifurcation structure is reduced, the point is determined as the final tail point, and the other point far from the point in step 6 is the final head point.

8. The non-destructive method for measuring body length of shrimp based on computer vision according to claim 3, wherein, In step 8, the centerline fitting method is as follows: from the head point to the tail point, the complete contour is divided into two sub-contours according to the clockwise and counterclockwise paths, the arc lengths of the two parts of the contour are calculated and are respectively subjected to equi-arc length sampling, the corresponding sampling points are obtained, the connecting lines of each pair of sampling points are calculated, and the midpoints of the connecting lines are calculated, finally, all the midpoints are sequentially connected by a smoothing curve, and thus a complete and continuous contour centerline is obtained; after the pixel length of the body length curve is calculated, the real body length of the prawn is obtained by multiplying the pixel length by a proportionality coefficient, wherein the proportionality coefficient is a built-in parameter of the camera.

Images