Study Of Freshwater Fish Species Identification And Online Weignt Detection And Equipment Design

Freshwater fish is a kind of nutritious ingredients and loved by most people. Freshwater fish have the disadvantage of perishable, difficult to store, etc. Its intensive processing can not only solve this problem, but also can enrich the form of fish consumption, such as canned fish, spices and other snack foods. Species and weight classification for freshwater fish is significant. In this paper, the study is on silver carp, grass carp, bream and crucian, four kinds of freshwater fish. The variety and the weight online grading detection methods are studied in view of the situation of no specialized freshwater fish’s variety and weight online grading equipment currently. Also, the grade prototyping and testing were completed. The paper concluded that:1) Firstly, a freshwater fish varieties classification method based on contour characteristic curve was proposed. Then the varieties classification model was established. By examining the variety grading model based on the contour characteristic of fish, the accuracy of the model’s grading on the four kinds of freshwater fish all reached 100%, which means the variety grading method has higher feasibility. The results also showed that the method’s recognition speed is faster compared to the other classification method, and the average recognition time is 1.32 s.2) On the base of the completion of freshwater fish’s variety and weight online grading detection methods’ research, a grading system was devised. The belt conveyor is a main body of the grading device. And both sides of the conveyor were installed with fish distribution mechanism. Online weighing device tests showed that with the increase of the running speed of the weighing conveyor, the online weighing average error of different weight group also increases. When weighing conveyor at a speed from 0.16m/s to 0.24m/s, the error range of the weighing device was ± 0.73%, which meets the technological requirements of freshwater fish’s preprocessor.3) By means of the simulation analysis of the grading device distribution mechanism rocker, this article selected 0.55 m as the arm length in order to ensure the smooth operation of the classifier at a higher speed, while reducing the impact of fish on rocker to extend the life of the steering gear. Meanwhile, in order to reduce the impact of the rocker on the steering shaft, the root of the rocker was designed as a shape of a certain arc transition. In case of ensuring the efficiency and accuracy of the grading device, the speed of the conveyor grading device could be set from 0.20m/s to 0.25m/s.4) The grading device was combined with the machine vision system and the online weighing device respectively, which builds the variety and weight online grading inspection prototype. The test results of the variety online grading inspection prototype showed that the accuracy of the prototype was 100% towards the white silver carp, grass carp and bream while the carp was 98%. The rate of grading was 13 per minute. The error between computer measurement value and manual measurement value was less than 20% when the eigenvalue was set higher than 85%.5) The test results of the weight online grading inspection prototype showed that the average speed of grading was 22 per minute and the accuracy was higher than 98% when the speed of the weighing equipment was set as 0.2m/s.