Cross-perception information interaction sensing fusion method in intelligent bionic evaluation for food

Abstract

The invention provides a cross-perception information interaction sensing fusion method in intelligent bionic evaluation for food quality. A huge matrix, comprising sensing information with different physical significances and different dimensions / magnitudes, is obtained by different bionic sensors. According to principal component analysis, all original characteristic variables are recombined to obtain a plurality of orthogonal principal components (PCi), so that the interaction sensing fusion of different types of sensing information is realized. The visual, olfactory and gustatory scores of manual organoleptic inspection are regressed according to the plurality of principal components so as to respectively obtain corresponding visual, olfactory and gustatory dummy variables. The dummy variables serve as the input of a decision making system, the nonlinear decision making system is constructed according to a genetic neural network method, the cross-perception sensor information interaction sensing fusion of visual, olfactory, gustatory and other sensations is truly realized, and a final comprehensive evaluation result approaches to human perception behaviors to the greatest extent.

Description

technical field [0001] The patent of the present invention relates to a method of cross-sensory bionic sensor information interaction induction fusion, which uses different types of sensors such as vision, smell, and taste to simulate human perception organs such as eyes, nose, and tongue to make a reasonable and accurate comprehensive evaluation of food quality. Background technique [0002] Human beings obtain food not only to meet physiological needs, but also to meet psychological needs and spiritual enjoyment. Foods with good or unique flavors not only directly increase appetite, promote the secretion of digestive juices, and strengthen digestion and absorption, but also make people feel happy in the senses. This is where the charm of food flavor lies. Food flavor refers to the stimulation of the human body's eyes, nose, tongue and other sensory organs before and after the food is eaten, thus causing people's comprehensive impression of its overall characteristics, whic...

AI Technical Summary

Problems solved by technology

For a long time, the inspection and evaluation of food flavor has usually relied on artificial sensory inspection, but artificial sensory inspection itself has certain limitations: 1) Artificial sensory inspection requires a long-term training and experience accumulation process, and generally relies on well-trained experts. Complete, ordinary consumers do not have this ability; 2) Artificial sensory testing is easily disturbed by external factors, even for well-trained experts, the sensitivity of their sensory organs will be affected by gender, experience, mental state, physical condition, The interference of factors such as motivation, attitude and regional environment affects the accuracy and consistency of the evaluation results

[0004] From the retrieval of papers and patents, it can be seen that currently in the intelligent evaluation of food quality, most of the bionic sensors use commercial instruments, which only play the role of analytical instruments, and the detection mechanism is still far from the artificial sensory inspection mechanism. There is no real "perception function" similar to that of human beings, and from published papers, journals and authorized patents, it can be known that some people have applied multi-information fusion technology to the non-destructive testing of food, agricultural products and other quality, such as "Intelligent evaluation method of famous tea quality instrument based on multi-sensor information fusion" (ZL200910232916.9) and "A food sensory quality evaluation method and system" (ZL201110142990.9), but in the fusion of multiple sensor information , it can be found that several sensor technologies are basically simply connected in series, and the sensory information with a large span of physical meaning is arbitrarily superimposed. In fact, this is not fusion in the true sense, and it is far from human perception behavior.

Images