Research And Application Of Beak Morphological Features Of Cephalopod Based On Computer Vision

The beaks are widely concerned in the study of cephalopods because of its characteristics of resistance to corrosion and good stability and contains a lot of biological information.Its application scope mainly includes species identification,population discrimination,feeding ecology,etc.Obtaining the morphological parameters of the beaks is the basis for the development of cephalopod biology.In traditional measurement methods,there are many samples,a large workload,and large accidental errors caused by humans,which reduce the speed of data collection and slow down the process of subsequent research.Computer vision technology can effectively solve this problem.As a new type of image processing technology,its advantage lies in the use of computers to convert the resulting images into digital images,through computer technology to simulate human judgment criteria,to identify the images,and merge with image analysis technology to analyze and draw the desired conclusions.In the fields of biology and fishery science,there is a great panorama of applications and development.Therefore,this study used the data of the calamari collected during the production of light purse seine in the Indian Ocean from January to February 2019 at 17°4′N—17°18′N,61°5′E—61°35′E.beak,using the self-made cephalopod beak multi-vision camera to take pictures of the horny jaw,and using computer vision technology to program the beak to select the morphological feature points and measure the morphological parameters and the morphological parameters of the manual radial measurement To compare,analyze and study the feasibility of computer vision in the study of cephalopod horny jaws,to lay the foundation for the realization of automatic measurement of beak parameters,and finally use the automatically measured morphological parameters to distinguish cephalopod types and populations.The main findings are as follows:(1)Design and manufacture of multi-vision rapid shooting device for jaws of cephalopods.The device is mainly divided into upper and lower parts.The upper part of the device is placed on the horny jaw,and the three views of the horny jaw are collected in one shot using the light reflection principle of the prism,and the three views of the horny jaw are displayed on the display on the side wall of the device.View,contour and feature point coordinates,extract the morphological parameters of the horny jaw;the lower part is the camera and the interface that can be connected to external devices,which can be used offline and online.(2)Contour and feature point extraction of on beak of cephalopods.In this study,we used computer vision to extract the contour and feature points of cephalopod beak.First,a self-made device was designed and programmed in MATLAB software,and used for capturing the three-dimensional view of the beak,and then the contour of the beak was extracted in the Canny algorithm.Finally,the feature point position was calibrated according to the definition of landmark,and the feature point coordinates of the beak were obtained by establishing the spatial coordinate system.The research results show that it was feasible to extract the contour image and feature point coordinates of the beak by computer visioning.When the σ value was 0.1,the contour image of beak had the best effect.The feature points were calibrated on the contour picture,and the spatial coordinates of each feature point were obtained by iterating through the contour picture.(3)Morphological parameter measurement on beak of cephalopods.First,the beak feature points and spatial coordinates were extracted using MATLAB.Next,the spatial distance between the feature points was calculated.Finally,the values of extracted beak morphological parameter were compared with the measured result.The results indicate that the arithmetic mean values of the 10 measurements obtained by the two methods were very close.The average absolute error,average relative error,standard deviation,and dispersion coefficient of computer vision measurements were less than manual measurements results obtained,except for the morphological parameter of the upper crest length.The arithmetic mean of the morphological parameters obtained from 10 repeated measurements of each cephalopod beak sample using two methods was close,but the average absolute error and average relative error of the data measured by computer vision were less than the average absolute error of the manually measured data.In addition,the average relative error indicates that the measurement results of computer vision were accurate and closer to the true value.Analysis of the standard deviation and dispersion coefficient shows that computer vision could repeatedly extract the morphological parameters of the beak of each sample multiple times.It was more concentrated near the true value and the precision was greater.(4)Discrimination of species and populations of Sepia.The horny jaw multi-vision automatic shooting device was used to capture the three views of the horny jaw and the determination of the morphological parameters of the squid captured in the Northwest Indian Ocean,the Middle East Pacific,and the East Indian Ocean.After the morphological parameters were standardized,a stepwise discriminant analysis was performed for population discrimination.The results showed that the average morphological parameters of the upper jaw and lower jaw of the squid in the Northwest Indian Ocean were larger than those of the other two sea areas;the morphological parameters of the horny jaws of the squid in the Central and Eastern Pacific and the East Indian Ocean were similar.There are eight effective variables in the discrimination of Squid species in the three sea areas: upper side wall length/carcass length ULWL/ML,lower ridge length/carcass length LCL/ML,upper beak length/carcass length URL/ML,lower Wing length/carcass length LWL/ML,upper wing length/carcass length UWL/ML,lower head cover length/carcass length LHL/ML,lower beak length/carcass length LRL/ML,upper head cover length/carcass length UHL/ML In the nine-step discrimination,the step that contributes the most is inputting lower beak length/carcass length LRL/ML,upper head cover length/carcass length UHL/ML,and the success rate of discrimination is 100%.In summary,the self-made cephalopod beak multi-vision camera is used to take pictures of the beak,and computer vision technology is used to program the beak for image contour extraction and morphological feature points selection,and the measurement of morphological parameters and manual path Comparing the measured morphological parameters,the automatic measurement of beak parameters is realized,and finally the cephalopod type and population are distinguished by using the automatically measured morphological parameters.The research results can provide a more reliable technical support for the measurement of cephalopod beak morphological parameters and the discrimination of cephalopod species and population.