348 results about "Food quality" patented technology

The invention discloses a food detection system and method based on computer vision, and relates to the technical field of food visual detection and intelligent identification, and the system comprises an image collection module which obtains initial food image data from a production line through a preset image collection device, processes the data through an edge detection algorithm to extract contour features, and sends the contour features to an image processing module; meanwhile, a boundary point coordinate extraction process is fused to obtain appearance boundary description of the food; the boundary clustering module is used for grouping boundary points by adopting a clustering algorithm according to the extracted appearance boundary description, carrying out processing under an iterative convergence condition in combination with a distance metric function and cluster center initialization, and judging that the shape is irregular if the grouping number exceeds a preset threshold value, so as to obtain a shape classification label; according to the food detection system and method based on computer vision, the accuracy and adaptability of food quality detection are remarkably improved, and the quality control requirement of a multi-scene production line is met.

The invention relates to the technical field of machine vision and food quality detection, in particular to a food defect real-time detection method and system based on image processing. The method specifically comprises the following steps: integrating multi-modal data, and adaptively adjusting a weight coefficient to improve data quality; multi-stage noise reduction is carried out, key frames are extracted, and a key frame verification mechanism is enhanced through a 2FA control variable; constructing a YOLO-BioNet model, and optimizing the detection efficiency and precision in a complex scene through dynamic convolution kernel allocation and a spiking neural network; integrating Kalman filtering, LSTM (Long Short Term Memory) time sequence modeling and a deep reinforcement learning model, analyzing a food surface change track and predicting a defect state; real-time data and historical information are fused, and a dynamic threshold value is calculated through a Bayesian network; performing linkage response between an automatic sorting instruction and a production line; a distributed storage and graph neural network optimization model is adopted, and defect tracing is supported. The method has remarkable advantages in the aspects of real-time performance, detection accuracy and self-adaptive capability.

The invention relates to the technical field of food quality detection, in particular to a tea oil authenticity identification method based on an in-situ mass spectrum-machine learning coupling technology. Comprising the following steps: S1, preparing a sample; s2, acquiring mass spectrum data; s3, feature extraction: performing dynamic binning processing on the mass spectrum data, extracting feature peak information, and performing time dimension feature aggregation to form an initial feature matrix; s4, data preprocessing; s5, training a machine learning model; and S6, intelligent identification and visual output: inputting the mass spectrum original data of the tea oil sample to be detected into the trained multi-model integrated system, outputting an identification result through weighted voting or an optimal model strategy, and giving a confidence interval to quantify the reliability of the result.

The invention relates to the technical field of electrochemical detection and food quality safety, and particularly discloses an edible oil rancidity degree rapid detection system based on an electrochemical sensor, comprising: a data acquisition module for receiving an original electrochemical response signal and performing format standardization, abnormal data rejection and integrity verification; the data preprocessing module adopts a statistical optimization algorithm to reduce noise, correct baseline drift, generate a unified dimension data set and complete stability verification; the feature analysis module is used for extracting rancidity related feature parameters through statistical correlation analysis, and carrying out fusion calculation after significance verification to obtain feature values representing rancidity; and the evaluation and judgment module inputs the characteristic value into a pre-training model, outputs a quantitative index through multivariate statistical calculation, dynamically sets a safety threshold in combination with the edible oil type, and outputs a rancidity judgment conclusion after comparison. The method depends on statistical data processing and analysis, the detection reliability and accuracy are improved, and the method is suitable for edible oil quality safety rapid screening and quality monitoring.

The invention discloses an intelligent food transportation monitoring method and system based on a multi-dimensional dynamic coupling algorithm, and the method comprises the steps: collecting food transportation data in real time through wireless communication; constructing a digital twinborn model of the food transportation process, and fusing the digital twinborn model and the risk prediction model; simulating a transportation environment change trend to obtain a current multi-dimensional dynamic coupling environment influence factor; obtaining a food heat accumulation amount, and predicting a food quality deterioration probability factor; judging whether the food quality deterioration probability factor exceeds a threshold value or not, and predicting a transportation risk multi-dimensional quantitative index by combining the multi-dimensional dynamic coupling environment influence factor and the food quality deterioration probability factor; generating an optimization strategy, and outputting an intervention scheme; according to the method, the defect of the traditional technology in the aspect of risk prediction of food transportation monitoring is overcome, and through accurate risk prediction and timely intervention measures, the deterioration loss of food in the transportation process is remarkably reduced, the transportation resource allocation is optimized, and the transportation efficiency is improved.

The invention discloses a multi-modal fusion-based roast chicken eating quality grading intelligent judgment method, which comprises the following steps of: performing matrix setting on a roasting process, preparing roast chickens with different quality grades, evaluating sensory attributes and overall scores of the roast chickens, and quantifying different modal data characteristics by adopting multiple quality detection methods. By constructing a process-sensory-multi-modal feature full-link database and combining a random forest for determining optimization and an XGBoost interpretable machine learning model, accurate grading and intelligent prediction of the roast chicken eating quality are realized, and the core problems of high subjectivity, poor single-modal generalization ability, data islands and the like of a traditional method are solved. Quality accurate grading prediction and digital traceability from processing to finished products are realized, and data support is provided for process optimization and market positioning prediction. The method has remarkable practical application value in the aspects of product comprehensive grade prediction and data management, and has prediction potential suitable for multiple scenes.

The invention discloses a food quality safety tracking management system based on big data, and relates to the technical field of food, and the system comprises the steps: collecting the label identification data and storage stacking parameters of to-be-tracked food; extracting a label risk coefficient between a label inconsistency degree and a quality compliance risk based on the historical supervision data, and processing the label risk coefficient and the label identification data to obtain a label risk value; according to the storage stacking parameters of the to-be-tracked food, the stacking abnormal degree value of the to-be-tracked food in the storage process is judged; processing the stacking abnormal degree value and the label identification data to obtain a storage standard interference value; according to the storage specification interference value, the storage specification interference degree of the to-be-tracked food is judged, and first-grade storage specification interference and second-grade storage specification interference are obtained; the method has the effect of improving the efficiency and effect of food quality safety management.

The invention relates to the technical field of food quality safety, and discloses a food quality safety traceability method based on big data, which comprises the following steps: step 1, binding a unique batch identification code for each food batch, and deploying Internet of Things sensing equipment to collect environmental parameters and geographical location information, after the environment parameters and the geographical location information are associated with the batch identification codes, a data packet with an identity label is formed; and 2, the cloud big data platform receives and gathers the data packets with the identity labels, integrates preset static basic data and node event data, and constructs a dynamic digital file for recording a complete life cycle process by taking batch identification codes as main keys. According to the technical scheme of binding unique identification codes for food batches and deploying Internet of Things sensing equipment to automatically collect and associate uploaded data to a cloud platform for convergence and fusion, automatic and integrated management of whole-course data of a supply chain is realized.

The invention discloses a quality mining and evaluation method of a wolfberry characteristic flavor nourishment, and relates to the technical field of food quality analysis and sensory evaluation, and the method comprises the following steps: S1, carrying out non-targeted analysis on the wolfberry characteristic flavor nourishment by adopting headspace solid phase microextraction in combination with comprehensive two-dimensional gas chromatography-time-of-flight mass spectrometry, and obtaining original abundance data of the volatile compound. According to the quality mining and evaluation method for the Chinese wolfberry characteristic flavor nourishment, accurate identification and quantitative evaluation of key flavor substances in the Chinese wolfberry characteristic flavor nourishment are realized by establishing a flavor contribution potential index model and combining a dual verification mechanism. According to the method, the limitation that only a single chemical component is concerned or actual sensory experience is ignored is solved, and the accuracy and reliability of flavor component screening are improved by integrating multi-dimensional parameters such as instrument response intensity, human olfaction threshold and matrix release characteristics and through cross validation of a gas chromatography-olfaction technology and a standard substance addition experiment.

The invention discloses a low-temperature preservation method of a prepared meat product, and belongs to the technical field of food processing. In order to solve the technical problems that in the low-temperature storage process of the conditioned meat product, microorganism control is not thorough, secondary pollution and fat oxidation aggravation are caused by accumulation of high-oxygen conditioned packaging condensate water, and texture damage is caused by ultrahigh-pressure treatment, metal ion catalysis is blocked through phytic acid pretreatment, oxidation is inhibited through composite antioxidant vacuum rolling and kneading, and the quality of the conditioned meat product is improved. Gradient ultrahigh pressure destroys microbial structures in stages, citric acid and a D-ribose buffer solution block damaged thalli from repairing, a composite packaging bag is lined with a hygroscopic film and filled with high-oxygen mixed gas to maintain meat color and adsorb condensate water, and finally, the composite packaging bag is deeply frozen to form a microcrystalline structure and then transferred to an ice-temperature environment for storage. The method realizes microbial safety control under the condition of no chemical preservative, effectively maintains the cell integrity of meat products, reduces juice loss and fat oxidation, prolongs the shelf life and maintains the eating quality.

The invention relates to the technical field of food quality detection, and discloses a meat product comprehensive scoring method based on principal component analysis, and the method comprises the steps: firstly obtaining physicochemical index (such as protein and fat content) and sensory attribute (such as color saturation and elasticity index) data of multiple batches of samples, and carrying out the structural arrangement of the two types of data according to batches and index dimensions; respectively carrying out Z-score standardization, range normalization and other processing, and extracting a physicochemical index feature vector and a sensory attribute feature vector; constructing a dynamic weight strategy through a variance contribution rate and a feature root ratio, and performing weighted fusion on the two types of feature vectors to obtain a fused feature vector; and finally, inputting the data into a scoring model, and after centesimal system conversion and distribution calibration, dividing quality grades in combination with an industry standard threshold to generate a comprehensive scoring result containing alphanumeric identifiers. The method realizes multi-dimensional data fusion and dynamic empowerment, improves the scoring accuracy, and is suitable for meat product quality evaluation.

The invention relates to a food multi-modal detection data fusion analysis method and device, equipment and a medium. The method comprises the following steps: acquiring original multi-source heterogeneous data of a food spectrum, an image and a smell, and performing normalization and noise reduction preprocessing to obtain a standardized feature set; classifying and screening high-correlation features by using a support vector machine, and generating a structured description; dynamic parameters are extracted, a quality trend vector is generated through time sequence analysis, a safety score is calculated, and a qualified mark is output if the safety score reaches the standard; constructing an extension index set based on the score, inputting a dynamic model to generate an authentication update link, and combining the block chain time to obtain new-version authentication; and associating the historical records to generate an initial report, and verifying and correcting to obtain an optimized report. By adopting the method, the consistency and availability of food safety data can be improved, the credibility of safety certification is enhanced, and a systematic and efficient solution is provided for food quality safety detection.

The invention discloses a grain supply chain privacy data sharing method based on distributed federated learning, and relates to the field of grain quality safety. According to the method, the private data sharing architecture of the grain supply chain is divided into a local layer, an edge layer and a global layer, a feature distillation framework is designed for data isomerism, and a model aggregation scheme and a hierarchical federated learning algorithm of each layer are optimized. And meanwhile, five encryption algorithms are introduced to realize hierarchical encryption, so that the data are ensured to be available but invisible, the sharing security and credibility are improved, and privacy protection and data sharing are balanced. Coupling of different types of data among links of the grain supply chain is increased, convenience and safety of data interaction among the links of the grain supply chain are improved, storage cost of the data of the supply chain and high delay of interaction are reduced, fine management of the data of the grain supply chain and personnel is achieved, and the management efficiency of the grain supply chain is improved. And the food quality safety of the grains is ensured.

The utility model belongs to the field of steam engines, particularly relates to a split type steam microwave heating machine suitable for boxed meal, and provides the following scheme aiming at the problems that the efficiency is low, the energy consumption is high and the taste and the nutritional value of food are damaged due to the fact that the existing equipment cannot quickly and uniformly heat the boxed meal, the split type steam microwave heating machine comprises a machine body, a machine door is hinged to one side of the machine body through hinges, a touch display screen is embedded in one side of the machine door, a frame is fixedly arranged in the machine body, a steam generator is fixedly arranged at the position, above the frame, in the machine body, and a water tank is fixedly arranged at the position, beside the steam generator, in the machine body. The water tank is communicated with the input end of the steam generator, and the output end of the steam generator is fixedly communicated with a steam spraying pipe. Through the arrangement of the steam generator and the steam spray pipe, the box lunch is quickly and uniformly heated, the quality and taste of food are ensured, meanwhile, the heating efficiency is improved, and the energy consumption is reduced.

The invention provides a food quality analysis system and method based on artificial intelligence. Belongs to the technical field of food detection and artificial intelligence. The system comprises a multi-modal data acquisition module used for synchronously acquiring a hyperspectral image, an odor signal and three-dimensional morphology data of food; the data processing and feature fusion module dynamically distributes proper weights for different modal features through an attention network by adopting a self-adaptive weighted feature fusion mechanism, and generates a fusion feature vector; the multi-task deep learning model module is used for simultaneously outputting prediction results of a plurality of quality indexes such as chemical components, sensory scores and defect classification of the food on the basis of the fusion features; and the quality evaluation and decision output module is used for carrying out comprehensive grading according to a prediction result. The method overcomes the defects that in the prior art, the detection dimension is single, and internal quality and sensory attributes cannot be associated, and achieves rapid, lossless, comprehensive and high-precision intelligent comprehensive analysis of food quality.

The utility model discloses a food quality detection sampling device relates to sampling device technical field, and it is including: sampling outer tube, and its one end is equipped with puncture part, sampling inner tube, and the inside of sampling outer tube is movably inserted, and with sampling outer tube rotatable, and the one end of sampling inner tube is equipped with the discharge gate in sampling outer tube outside, handheld part, set up in sampling outer tube and sampling inner tube away from the one end of puncture part, outer hole, open in the circumferential side of sampling outer tube, inner hole, open in the circumferential side of sampling inner tube, reinforcing pad, fixed in the inside of sampling inner tube, and located the opposite side of inner hole, through the inside fixed setting reinforcing pad in the opposite side of inner hole, the bending strength that significantly made up for the loss because of the trepanning, prevent sampling inner tube from bending deformation in the process of insertion or rotation, improved the structural stability of device under high pressure food detection and other working conditions.

Disclosed are a nano colorimetric sensor based on natural pigment signal enhancement and a preparation method thereof, the nano colorimetric sensor being a nano colorimetric sensor array formed by nano colorimetric materials, the nano colorimetric materials comprising POEs materials and natural pigments. After 6 POEs materials and 4 natural pigments are respectively self-assembled under magnetic stirring, the nano colorimetric materials with sensitization effect are screened through RGB response values and Euclidean distances, the precursor substance mass ratio is optimized, and finally the sensitization effect is verified. The nano colorimetric sensor has high sensitivity, the colorimetric material is green and environmentally friendly, and can effectively realize rapid quantitative detection of the freshness of aquatic products, and has important practical significance in meeting the needs of consumers for food quality and safety and maintaining market order.

The invention discloses a honey walnut kernel raw material two-dimensional quality screening system, and relates to the technical field of food quality detection, and the system comprises a raw material reference database module; a raw material two-dimensional detection module; a dynamic parameter calibration module; a multi-dimensional quality determination module; a reference data system of multiple production areas and multiple seasons is established through the raw material reference database module, basic indexes and interface interaction data are synchronously obtained in cooperation with the raw material two-dimensional detection module, and the problem that traditional detection only pays attention to the quality of a single raw material and ignores the combination quality of honey and walnut kernels is solved; meanwhile, the dynamic parameter calibration module calculates the comprehensive quality deviation and the equipment adjustment amount, self-adaptive optimization of detection parameters is achieved, insufficient precision caused by fixed parameters is avoided, and raw material quality abnormity can be recognized from multiple aspects of raw material basic attributes and interface interaction states by constructing a two-dimensional screening system of raw material basic quality and interface combination degree.

The utility model relates to a food quality detection device which comprises a first path, a second path and a third path, and an object to be detected passes through the first path, the second path and the third path in sequence; the first path is provided with a detection part, and the second path is in linkage fit with the detection part and forms releasing or resetting. Through cooperation of the three paths, after the packaging bags are scanned and detected, the unqualified packaging bags can fall off from the middle of the conveying line, and the qualified packaging bags are directly conveyed to the tail end of the conveying line, so that automatic sorting of the qualified packaging bags and the unqualified packaging bags is completed; compared with the prior art, the detection efficiency can be effectively improved.

The invention discloses a method for magnetic field assisted vacuum cooling of a high-temperature flour product and application thereof, and the method comprises the following steps: the flour product is placed in a magnetic field assisted vacuum cooling environment for cooling treatment, the magnetic field is a static magnetic field, the magnetic field intensity is set to be 0.1-5T, the vacuum degree is set to be 0.1-2kPa, and in the cooling treatment process, the magnetic field intensity is set to be 0.1-5T, and the vacuum degree is set to be 0.1-2kPa; and the center temperature of the flour product is reduced from 50-100 DEG C to 0-30 DEG C. According to the magnetic field assisted vacuum cooling method, the problems of water loss and mass loss in the vacuum cooling process of the high-temperature flour products are solved, the food quality is improved, and the method is easy to operate, low in cost and remarkable in benefit and shows wide application prospects in the field of precooling and preservation of the high-temperature baked or cooked flour products.

The utility model provides an ohmic unfreezing electrode plate and a refrigerator drawer with the ohmic unfreezing electrode plate, which comprise two electrode plate assemblies arranged at an interval and an insulating module arranged between the electrode plate assemblies, the insulating module is used for separating the two electrode plate assemblies; the two electrode plate assemblies comprise conductive materials and are respectively connected with opposite electrodes; and the two electrode plate assemblies and the insulating module form an integral flat-plate-shaped structure. According to the ohmic unfreezing electrode plate and the refrigerator drawer with the ohmic unfreezing plate, the application scenarios of the ohmic unfreezing electrode plate are increased, the surface area of the ohmic unfreezing electrode plate is increased by arranging the concave-convex parts on the upper surface of the electrode plate assembly, then the contact area with food is reduced, the area of the contact position of the electrode plate assembly and the food is reduced, the heat conduction speed is high, and the unfreezing effect is good. The heat at the contact part is reduced, and the damage to the food quality is avoided.

The utility model discloses an anti-breaking device for rice processing, which belongs to the technical field of rice processing and comprises a rice processing cabin, a plurality of pipe chain conveyors are assembled on the rice processing cabin, and each pipe chain conveyor is provided with a conveying pipe and a driving mechanism. According to the anti-breaking device for rice processing, the scraping pieces drive rice between the two scraping pieces to be conveyed synchronously in the process that the scraping pieces slide in the conveying pipe along with the chain, meanwhile, the grain storage space in the corner driven wheel cavity is reduced through attachment of the corner driven wheel cavity and the driven wheel, and therefore the grain storage efficiency is improved. According to the rice conveying device, excessive rice is prevented from being accumulated in a corner driven wheel cavity to be collided and rubbed, friction between the rice and the inner wall of a conveying pipe in the conveying process of the rice in the conveying pipe is reduced, the risk that the rice is broken due to collision, extrusion and friction in the conveying process is further reduced, the integrity and physical characteristics of rice particles are ensured, and the rice conveying efficiency is improved. The economic value, the storage processing performance and the eating quality of the rice are further improved.

The application belongs to the technical field of food quality monitoring, and particularly relates to a ferric magnetic solid-phase masking agent and a quantitative detection method for lead in vegetable products. The ferric magnetic solid-phase masking agent is prepared by the following method: S1, performing carboxylation on the terminal amino group of DFO to obtain DFO-COOH; S2, coating a SiO2 layer on the surface of Fe3O4 magnetic core to obtain Fe3O4@SiO2; S3, modifying the Fe3O4@SiO2 by amination to obtain Fe3O4@SiO2-NH2; and S4, coupling Ac-DFO with Fe3O4@SiO2-NH2 to obtain Fe3O4@SiO2-DFO. The three hydroxamic acid groups of DFO are combined with Fe 3+ to form a stable chelate, and Fe 3+ can be removed through magnetic separation, the removal rate is increased to more than 99%, and the quantitative detection of lead ions is further faster and more accurate in combination with mature immunochromatography technology.

The invention discloses vinasse bread premixed flour and a preparation method thereof, and belongs to the technical field of bread premixed flour. The premixed flour comprises the following components in percentage by weight: 40%-60% of vinasse powder, 30%-40% of high gluten wheat flour, 5%-15% of rye flour, 3%-8% of vital gluten, 0.5%-1.5% of edible corn starch and 1%-2% of a bread improver. The bread improver is prepared by compounding mono-and diglycerides of fatty acids, diacetyl tartaric acid ester of mono-and diglycerides, a plurality of enzyme preparations and vitamin C. The preparation method comprises the steps of raw material pretreatment, mixing and the like. Compared with the prior art, by means of an optimized formula system, the technical problems that a dough system is degraded, the taste is hard, and aging is prone to occurring due to the fact that vinasse is added in a high proportion are successfully solved, and the finally-made bread is softer in taste and rich in elasticity, prolongs the shelf life and has high nutritive value and excellent eating quality.

This invention belongs to the field of biomaterial preparation technology, specifically disclosing a self-assembled supramolecular peptide nanozyme material, its preparation method, and its application. The preparation method includes the following steps: S1, dissolving β-alanyl-L-histidine in ultrapure water and stirring until homogeneous to obtain solution A; S2, dissolving heme chloride in dimethyl sulfoxide and stirring until homogeneous to obtain solution B; S3, mixing solution A and solution B until homogeneous, adjusting the pH, heating in a water bath, incubating at room temperature, centrifuging, discarding the precipitate, and retaining the supernatant; S4, lyophilizing and grinding the supernatant obtained in S3 to obtain the self-assembled supramolecular peptide nanozyme material. This invention discloses a self-assembled supramolecular peptide nanozyme material, its preparation method, and its application. This supramolecular peptide nanozyme, when applied to food quality and safety testing, can achieve visual and sensitive colorimetric detection of biogenic amines in spoiled food.

The utility model relates to a food metal detector. Comprising two first conveying belts arranged at intervals and a second conveying belt located between the two first conveying belts, and the conveying direction between the second conveying belts is opposite to the conveying direction of the first conveying belts; the metal detector body is located above the first conveying belt and the second conveying belt. Guide plates for guiding food to the second conveying belts are arranged at one ends of the two first conveying belts; baffles are arranged between the two first conveying belts and the second conveying belt. When food is conveyed, the food is firstly conveyed through the first conveying belt and passes through the metal detector body for the first time, when the food reaches the guide plate through the first conveying belt, the food collides with the guide plate, the food enters the first conveying belt under the action of the guide plate, and in the whole process, the angle of the food changes immediately; and then the food is re-inspected on the second conveying belt through the metal inspection machine body, so that the detection error caused by the angle of the food is effectively avoided, and the detection effect of the metal inspection machine is further improved.

The utility model discloses a sampling device for food quality detection belongs to food detection technical field, and its technical scheme main points include sampling cylinder, the top of sampling cylinder is provided with the top cover, the bottom of top cover is provided with cutting mechanism, the bottom of sampling cylinder is provided with the bottom cover, the bottom of sampling cylinder is provided with support mechanism, the cutting mechanism includes pressing block, spring, sliding rod and cutting cylinder, the bottom of pressing block is fixedly connected with the top of sliding rod, the sliding rod slidingly inserts inside the top cover, the surface of spring is movably sleeved in the sliding rod, the top of spring is in contact with the bottom of pressing block, the bottom of spring is in contact with the top of top cover, the structure of the food sampling of existing part is solved, especially when the quality of vegetable food is detected, it is more inconvenient, still needs manual cutting sampling with blade, and the efficiency is lower, and the problem that sample pollution is easily caused is solved.

The invention relates to the technical field of food quality inspection, in particular to a multi-dimensional analysis food quality detection system, which comprises a data acquisition module for performing non-destructive physical detection on a target food sample, acquiring a multi-source sensing signal and generating a food original physical characteristic data stream; the feature extraction module is used for constructing a multi-dimensional physical feature vector matrix of food and carrying out dimension standardization and redundant feature elimination processing; the characteristic correlation analysis module is used for performing similarity comparison on the multi-dimensional physical characteristic vector matrix and a preset food component database, outputting a food component estimation result and establishing a mapping model between physical characteristics and chemical indexes; and the quality evaluation module is used for comparing a food component estimation result with a food standard parameter, generating a quality risk grade evaluation report and dynamically adjusting a sensing strategy. According to the invention, the safety, efficiency and broad-spectrum adaptive capacity of food detection are improved, and the method is especially suitable for online rapid screening scenes of complete commodities, raw materials or in-transit food.

The application discloses a food full-link quality deterioration tracing method and system, a medium and a product, and relates to the technical field of food quality tracing.The method is implemented, and a sensory description label in consumer feedback information is extracted by a tracing system; with the aid of a preset food quality knowledge graph, accurate mapping of subjective sensory information and objective physical and chemical indexes is realized, and the problem that subjective feedback and objective data are disconnected in a traditional tracing system is solved.The tracing system carries out dynamic deduction on food full-link monitoring data through a preset quality deterioration dynamics model, and upgrades food quality deterioration from static threshold judgment to dynamic numerical evolution track analysis, so that the cumulative deterioration effect of each circulation link can be quantified.The tracing system screens food quality deterioration responsibility links based on average deterioration rate and link deterioration increment, traces deterioration inducement of the responsibility links in combination with the preset food quality knowledge graph, accurately defines the deterioration responsibility and traces the reasons, and improves the tracing efficiency and accuracy.