CN-121981979-A - Meat freshness detection method, device, equipment, storage medium and product

CN121981979ACN 121981979 ACN121981979 ACN 121981979ACN-121981979-A

Abstract

The present disclosure relates to the field of food detection technologies, and in particular, to a method, a device, equipment, a storage medium, and a product for detecting freshness of meat, where the method includes acquiring smell signals corresponding to preset smells released by meat food materials at a plurality of times before a current time, and acquiring surface image information, environmental information, and storage time of the meat food materials at each time; determining the corresponding target weight of the smell signal, the surface image information, the environment information and the storage time length at each moment, and detecting the freshness index value of the meat food at the current moment according to the target weight, the smell signal, the surface image information, the environment information, the storage time length and a preset detection model. The method can solve the problems of inaccurate detection of the freshness of the meat and poor real-time property.

Inventors

ZHANG CHENYANG
ZHAO MING
ZHANG YIHANG
ZHU XIAODONG
ZHOU YI
ZHAO BOYANG
SHEN WENFENG

Assignees

中科微感(宁波)科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260113

Claims (10)

1. A method for detecting freshness of meat, the method comprising: acquiring smell signals corresponding to preset smells released by meat food materials at a plurality of moments before the current moment, and acquiring surface image information, environment information and storage time of the meat food materials at each moment; Determining the target weight corresponding to each smell signal, the surface image information, the environment information and the storage duration corresponding to each moment; and detecting the freshness index value of the meat food material at the current moment according to the target weight, the smell signal, the surface image information, the environment information, the storage duration and a preset detection model.
2. The method of claim 1, wherein the sensor array is composed of a plurality of sensors, the odor signal is a resistance value when each sensor contacts the preset odor at each time, the resistance value when each sensor contacts the preset odor at the same time is different, and the acquiring odor signals corresponding to the preset odor released by the meat food at a plurality of times before the current time comprises: Based on a sensing array, collecting real-time resistance values of each sensor at each moment when the sensors contact the preset smell; and obtaining smell signals corresponding to each of a plurality of moments according to the preset resistance weight value and the real-time resistance value corresponding to each sensor.
3. The method of claim 1, wherein determining the target weights for each of the scent signal, the surface image information, the environmental information, and the storage time period for each of the time instants comprises: determining initial weights corresponding to the smell signals, the surface image information, the environment information and the storage time periods respectively corresponding to the time points based on the storage time periods corresponding to the time points; And according to the surface image information or the smell signal, respectively adjusting initial weights corresponding to the smell signal, the surface image information, the environment information and the storage time according to a preset proportion to obtain target weights corresponding to the smell signal, the surface image information, the environment information and the storage time.
4. The method of claim 3, wherein determining the initial weight for each of the scent signal, the surface image information, the environmental information, and the storage time period for each of the time instants based on the storage time period for each of the time instants comprises: when the storage time length corresponding to each time is smaller than or equal to a first threshold value, setting an initial weight corresponding to the smell signal as a first weight, setting an initial weight corresponding to the surface image information as a second weight, setting an initial weight corresponding to the environment information as a third weight, setting an initial weight corresponding to the storage time length as a fourth weight, and setting the sum of the second weight, the third weight and the fourth weight to be smaller than the first weight; Or when the storage time length corresponding to each moment is greater than the first threshold and less than or equal to the second threshold, setting the initial weight corresponding to the smell signal as a fifth weight, the initial weight corresponding to the surface image information as a sixth weight, the initial weight corresponding to the environment information as a seventh weight, the initial weight corresponding to the storage time length as an eighth weight, and the sum of the fifth weight, the sixth weight and the seventh weight is smaller than the eighth weight; Or when the storage time length corresponding to each moment is greater than the second threshold, setting an initial weight corresponding to the smell signal as a ninth weight, wherein the initial weight corresponding to the surface image information is a tenth weight, the initial weight corresponding to the environment information is an eleventh weight, the initial weight corresponding to the storage time length is a twelfth weight, the ninth weight is greater than the tenth weight, the tenth weight is greater than the twelfth weight, and the twelfth weight is greater than the eleventh weight.
5. The method of claim 1, wherein detecting the freshness index value of the meat food material at the current time based on the target weight, the smell signal, the surface image information, the environmental information, the storage time period, and a preset detection model comprises: extracting characteristics of the smell signal, the surface image information, the environment information and the storage time length to obtain smell characteristics, image characteristics, environment characteristics and storage time length characteristics corresponding to each moment; And inputting the smell characteristics corresponding to each moment, the target weights corresponding to the smell signals, the image characteristics, the target weights corresponding to the surface image information, the environment characteristics, the target weights corresponding to the environment information, the storage duration characteristics and the target weights corresponding to the storage duration into the preset detection model, and detecting to obtain the freshness index value of the meat food material at the current moment.
6. The method according to claim 1, wherein the method further comprises: If the output freshness index value at the current moment is larger than the first index value, determining that the meat food material is fresh; or if the freshness index value at the current moment is smaller than or equal to the first index value and larger than the second index value, determining that the meat food material is in the secondary freshness level, and displaying prompt information, wherein the prompt information is used for prompting that the meat food material is in the secondary freshness level; or if the output freshness index value at the current moment is smaller than the second index value, determining that the meat food material is of a spoilage grade, and displaying alarm information, wherein the alarm information is used for prompting that the meat food material cannot be eaten.
7. The utility model provides a detection device of meat freshness, its characterized in that, detection device of meat freshness includes: The acquisition module is used for acquiring smell signals corresponding to preset smells released by the meat food materials at a plurality of moments before the current moment, and acquiring surface image information, environment information and storage time length of the meat food materials at each moment; The processing module is used for determining the target weight corresponding to each smell signal, the surface image information, the environment information and the storage time corresponding to each moment, and detecting the freshness index value of the meat food material at the current moment according to the target weight, the smell signal, the surface image information, the environment information, the storage time and a preset detection model.
8. A computer device, comprising: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the method of detecting freshness of meat as claimed in any one of claims 1 to 6.
9. A computer-readable storage medium having stored thereon computer instructions for causing a computer to perform the method of detecting freshness of meat as set forth in any one of claims 1 to 6.
10. A computer program product comprising computer instructions for causing a computer to perform the method of detecting freshness of meat as claimed in any one of claims 1 to 6.

Description

Meat freshness detection method, device, equipment, storage medium and product Technical Field The disclosure relates to the technical field of food detection, in particular to a method, a device, equipment, a storage medium and a product for detecting freshness of meat. Background As people pay more and more attention to food safety, detection of freshness of meat food materials becomes more and more important. Traditional meat food freshness detection generally depends on a single index (such as total volatile basic nitrogen, pH value, K value and total colony count), and the mode needs to sample and analyze the sample chemically, so that the method has destructiveness, long detection period, cannot be detected on line and cannot be used for real-time quality monitoring. In addition, traditional meat food freshness may also rely on sensory evaluation, such as color, smell and taste, which is subjective and not accurate enough. Disclosure of Invention In order to solve the technical problems, the present disclosure provides a detection of freshness of meat, so as to solve the problems of inaccurate detection and poor real-time performance of freshness of meat. The method comprises the steps of obtaining smell signals corresponding to preset smells released by meat food materials at a plurality of moments before the current moment, obtaining surface image information, environment information and storage duration of the meat food materials at each moment, determining target weights corresponding to the smell signals, the surface image information, the environment information and the storage duration of the meat food materials at each moment, and detecting freshness index values of the meat food materials at the current moment according to the target weights, the smell signals, the surface image information, the environment information and the storage duration and a preset detection model. Based on the method, odor signals corresponding to preset odors released by the meat food materials at a plurality of moments before the current moment can be obtained, surface image information, environment information and storage time length of the meat food materials at each moment are obtained, target weights corresponding to the odor signals, the surface image information, the environment information and the storage time length of each moment are determined, and freshness index values of the meat food materials at the current moment are detected according to the target weights, the odor signals, the surface image information, the environment information and the storage time length and a preset detection model. The multi-mode data of smell, images, environment and storage time length at a plurality of times before the current time are collected, so that the dynamic evolution process (such as gradual increase of smell concentration and slow darkness of image color) from freshness to deterioration of meat can be captured, one-sided judgment caused by only relying on single-time data is avoided, and multi-dimensional information is mutually verified, so that the misjudgment risk of single-mode information is reduced. In an alternative implementation mode, the sensing array is composed of multiple sensors, the smell signals are resistance values when each sensor contacts preset smell at each moment, the resistance values when each sensor contacts the preset smell at the same moment are different, the smell signals corresponding to the preset smell released by meat food materials at multiple moments before the current moment are obtained, the real-time resistance values when each sensor contacts the preset smell at each moment are collected based on the sensing array, and the smell signals corresponding to each moment in multiple moments are obtained according to the preset resistance weight values and the real-time resistance values corresponding to each sensor. In an alternative implementation mode, determining the target weights corresponding to the smell signal, the surface image information, the environment information and the storage time period respectively at each moment includes determining the initial weights corresponding to the smell signal, the surface image information, the environment information and the storage time period respectively at each moment based on the storage time period corresponding to each moment, and adjusting the initial weights corresponding to the smell signal, the surface image information, the environment information and the storage time period respectively according to the surface image information or the smell signal to obtain the target weights corresponding to the smell signal, the surface image information, the environment information and the storage time period respectively. In an alternative embodiment, based on the storage time length corresponding to each moment, determining the initial weight corresponding to each moment, including setting the initial weight corresponding to the