Starch particle identification method based on Mask R-CNN

Abstract

The invention discloses a Mask R-CNN-based starch particle identification method, and the method combines hot stage-polarized light microscopic observation with a computer image identification technology, and intelligently identifies the change of the number of particles in a starch gelatinization process: collecting starch gelatinization image data based on hot stage-polarized light microscopic observation; sequentially carrying out starch gelatinization data set construction, Mask R-CNN model construction and training, starch gelatinization process particle detection and starch particle number statistics through an algorithm, wherein the algorithm comprises a starch image binarization algorithm, a starch gelatinization data set construction algorithm, a neural network training algorithm, a starch particle detection algorithm and the like. According to the method, the Mask R-CNN is used for intelligently detecting starch gelatinization, the automation degree is high, the detection efficiency is good, and the particle detection accuracy rate reaches 95% or above. The invention provides a novel quantitative method for intelligently detecting starch particles and evaluating starch gelatinization.

Description

technical field [0001] The invention belongs to the technical field of food science and engineering, and in particular relates to a method for identifying starch granules based on Mask R-CNN. Background technique [0002] In the traditional food industry, starch is an important component that undertakes the structure and function of food. In recent years, starch has been widely used in food, material, chemical, medical and other fields. Gelatinization is an intuitive expression of component morphology, supramolecular structure, multi-component structure-activity relationship and behavioral response in starch systems. During processing, under the action of relevant physical fields, the intramolecular and intermolecular interactions of starch molecules (chains) are weakened, resulting in changes in multi-scale structures (granule structure, ultrastructure, crystal structure and molecular structure), specifically manifested as swelling, Phase transition behaviors such as gela...

AI Technical Summary

Problems solved by technology

At present, most of the research on the number of granules in the starch gelatinization process is focused on manual recognition, which has a slow recognition rate and incomplete recognition indicators, which affect the gelatinization evaluation.

Due to the non-uniform granule evaluation standards, the edge shape, color and granule volume are all different, the number of starch gelatinized granules is difficult to quantify automatically under the current research method, which limits further research on starch gelatinization indicators

Images