Multi-sensing information fusion based instrumental intelligent evaluation method for quality of famous tea

Abstract

The invention provides a multi-sensing information fusion based instrumental intelligent evaluation method for the quality of famous tea, and relates to an instrumental intelligent evaluation method aiming at the quality of the famous tea. The method comprises the following steps: (1) aiming at a certain type of famous tea, selecting a batch of representative samples, performing sensory evaluation on various quality indexes of the famous tea by professional staff, and establishing a standard database; (2) respectively acquiring sensing information capable of reflecting the quality indexes of the famous tea with different sensors, and transmitting the sensing information to a computer; (3) extracting characteristic variables and respectively associating characteristic vectors with results of the sensory evaluation of experts in the database through the computer, and realizing identity description of the sensors to the quality indexes of the famous tea; and (4) simulating a human brain with the computer, establishing an instrumental intelligent evaluation multi-information fusion model for the famous tea through pattern recognition, and substituting the characteristic vectors extracted from the samples to be tested into the pre-established model to evaluate unknown samples. Compared with sensory detection of human, the method has good consistency of evaluation results and high automation degree.

Description

technical field [0001] The invention relates to an instrument intelligent evaluation method for the quality of famous tea, in particular to an instrument intelligent evaluation method for the quality of famous tea based on the fusion of multi-sensing information. Background technique [0002] Due to the high price of famous tea and the large profit margin of the product, to a certain extent, there is a phenomenon of shoddy and fake ones in the famous tea market. Famous and high-quality tea products are mixed up, and the quality of good and bad varies, which brings purchase risks to consumers, which not only damages the interests of consumers, but also is not conducive to the brand maintenance of famous and high-quality tea. For a long time, sensory evaluation has been regarded as the most basic method to test the quality of tea. The content of evaluation by well-trained professional reviewers generally includes color, aroma, taste, shape and leaf bottom (tea dregs), etc. Re...

AI Technical Summary

Problems solved by technology

However, sensory evaluation has the following two limitations: ① Sensory evaluation is an expert behavior after all, and the cultivation of experts is a long-term training and experience acquisition process. It takes a lot of effort to train a well-trained sensory evaluation expert. Ordinary consumers generally do not have this ability at a high price; ②The review is greatly disturbed by subjective and objective factors. Even for well-trained senior review experts, the sensitivity of their sensory organs is also affected by experience, gender, mental state, physical conditions and even the interference of factors such as the geographical environment, which will affect the accuracy and stability of the sensory evaluation results

According to the search results, there is currently only one domestic related patent (utility model) on computer vision identification of tea categories, "A device for identifying different green teas based on texture analysis based on multi-spectral image technology", patent number: ZL2007201100404.1; The application of electronic nose and electronic tongue technology in tea quality detection is still in the laboratory stage, and there are no relevant patent documents

And in the tea quality detection, what is adopted is a single detection method, which has certain limitations.

There are many indicators to reflect the quality of tea, including sensory indicators such as color, aroma, taste, shape, etc., and a single detection method often cannot fully describe an object, but can only describe one aspect of it. This kind of focus is different. The limitations brought about will inevitably affect the accuracy and stability of the test results

Images