Intelligent sensory evaluation detection method for cooking flavor of rice

Abstract

The invention discloses an intelligent sensory evaluation detection method for rice cooking flavor, and relates to the technical field of food detection. An electronic nose module, a texture analysis module, a machine vision module, a data processing module and a comprehensive evaluation module are adopted, optimization screening is carried out according to the characteristics of volatile components of rice, and sensors sensitive to key aroma substances such as aldehydes, esters and heterocycles of rice are included. By integrating the electronic nose, the texture analysis module and the machine vision module, the rice cooking flavor is accurately detected in multiple dimensions, the subjective influence is reduced, the evaluation objectivity is enhanced, all the modules are high in automation degree, detection is efficient and rapid, the market requirement can be met, meanwhile, multiple technologies are creatively fused, the related technology blank is filled, and the market prospect is wide. The intelligent evaluation model can be continuously optimized, various rice quality evaluation requirements are met, and the development of a food detection technology is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of food detection, and more particularly to an intelligent sensory evaluation detection method for the cooking flavor of rice. Background Art

[0002] As an important staple food globally, the cooking flavor quality of rice has attracted much attention. With the improvement of living standards, consumers' requirements for the flavor quality of rice are becoming more diverse. They not only pursue hygiene and nutrition but also pay more attention to appearance, palatability, and aroma. However, there are many problems in the current evaluation of rice flavor. Traditional sensory evaluation relies on the subjective judgment of evaluators, which is easily interfered by factors such as the fatigue degree and mood of evaluators, and there are differences among different evaluators, resulting in the lack of objectivity and stability of evaluation results. Although physical and chemical analysis can measure some indicators, it is cumbersome and time-consuming to operate, and it is difficult to meet the needs of rapid detection.

[0003] In terms of the detection of rice aroma, although it is known that a variety of volatile flavor substances affect rice aroma, such as 2-acetyl-1-pyrroline giving the sweet aroma of cooked rice and ester compounds enhancing the aroma, the existing technologies have limited means for the comprehensive detection and accurate evaluation of rice aroma, and the application of electronic nose technology in the analysis of rice aroma characteristics is also less. In the detection of texture characteristics, although it has been clarified that the contents of starch, moisture, and protein affect the taste of cooked rice, and a texture analyzer can measure relevant parameters, how to more comprehensively and efficiently use these parameters for quality evaluation remains to be improved. For the appearance evaluation of cooked rice, the research on the rapid detection of the taste quality of cooked rice by machine vision technology is still insufficient.

[0004] In summary, the existing technologies are difficult to accurately, quickly, and comprehensively evaluate the cooking flavor quality of rice, and cannot meet the market's demand for rice quality detection. There is an urgent need to develop an innovative intelligent sensory evaluation detection method. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent sensory evaluation detection method for the cooking flavor of rice to solve the problems existing in the above background art.

[0006] The present invention provides the following technical solutions: An intelligent sensory evaluation detection method for the cooking flavor of rice, including an electronic nose module, a texture analysis module, a machine vision module, a data processing module, and a comprehensive evaluation module: The electronic nose module selects an electronic nose device with high precision and multiple sensors. Its sensor array is optimized and screened according to the characteristics of rice volatile components, and includes sensors that are sensitive to key aroma substances in rice such as aldehydes, esters, and heterocycles, and is used for the rapid and accurate detection of the main aroma components of rice;

[0007] The texture analysis module uses a texture analyzer equipped with a high-precision mechanical sensor and a stable mechanical transmission device, and performs continuous compression tests on rice in the TPA mode to simulate oral chewing, so as to accurately measure texture parameters such as the hardness, elasticity, viscosity, chewiness, and resilience of rice;

[0008] The machine vision module consists of a high-resolution camera, professional image processing software, and an illumination system. The camera is used to collect images of cooked rice, and the image processing software is used to analyze appearance characteristics such as color, surface structure, and appearance shape in the image, and extract characteristic information such as CIE-L*, a*, b* color values, and particle shape parameters;

[0009] The data processing module uses algorithms such as principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) to perform preprocessing, feature extraction, and dimensionality reduction on the data collected by the electronic nose module, texture analysis module, and machine vision module;

[0010] The comprehensive evaluation module, based on multi-sensor fusion technology, fuses the data obtained from the electronic nose, texture analysis, and machine vision, establishes intelligent evaluation models such as artificial neural network (ANN) and hierarchical cluster analysis (HCA), combines artificial sensory evaluation data for training and optimization, and realizes comprehensive and objective evaluation of the cooking flavor of rice and outputs evaluation results. Further, the optimization and screening of the sensor array in the electronic nose module is based on the GC-MS analysis results of the volatile components of different varieties of rice to determine the sensor combination that is sensitive to the key aroma substances of rice.

[0011] Further, when the texture analyzer in the texture analysis module performs TPA tests, the continuous compression test process simulating oral chewing conforms to the standard operation procedure to ensure the accuracy of texture parameter measurement.

[0012] Further, the illumination system of the machine vision module provides uniform and stable illumination conditions to ensure the quality of the rice images collected by the high-resolution camera, and facilitate the accurate extraction of appearance feature information by the image processing software.

[0013] Further, the preprocessing of the data collected by the data processing module includes operations such as noise removal and normalization to improve the data quality.

[0014] Further, during the training and optimization of the intelligent evaluation model established by the comprehensive evaluation module, the model parameters are continuously adjusted to improve the accuracy and generalization ability of the model for evaluating the cooking flavor of rice.

[0015] An intelligent sensory evaluation detection method for the cooking flavor of rice based on the above modules is carried out according to the following steps:

[0016] S1. Sample preparation: Collect paddy samples from multiple rice-producing areas, and process them through hulling and milling to obtain standard third-class precision rice that complies with GB1354-2009. Store the rice samples in a 2°C freezer; refer to GB / T15682-2008 and improve the preparation of cooked rice. Weigh 10.0 g of rice samples into a special aluminum box, wash them with distilled water according to the water-rice ratio of 1.3:1, soak them at 25°C for 30 min, steam them in a steamer for 40 min, and keep them warm and simmer for 20 min;

[0017] S2. Physicochemical index determination and manual sensory evaluation: Use gas chromatography-mass spectrometry (GC-MS) to analyze the odor of rice of different varieties during different cooking processes, and determine the content of the main aroma substances; adopt the quantitative description method, organize trained sensory evaluators to conduct manual sensory evaluation on the rice under different cooking methods, record the evaluation results, and construct a manual sensory evaluation data set;

[0018] S3. Multi-module detection: The electronic nose module detects the aroma of rice samples according to the optimized sensor array, obtains the sensor response signal and converts it into a digital signal for transmission to the data processing module; the texture analysis module conducts TPA tests on the cooked rice samples cooled to room temperature, and records various texture parameters; the machine vision module collects rice images, analyzes and extracts appearance feature data through image processing software, and transmits them to the data processing module together;

[0019] S4. Data processing and model construction: The data processing module processes the data transmitted by each module, uses algorithms such as PCA and PLS-DA to extract features and reduce the data dimension; the comprehensive evaluation module uses multi-sensor fusion technology, combines algorithms such as ANN and HCA to construct an intelligent sensory evaluation model, and inputs the processed data into the model for training and optimization;

[0020] S5. Evaluation and result output: Input the detection data of the rice samples to be detected into the optimized intelligent sensory evaluation model, and the model outputs the evaluation results of the cooking flavor of the rice, including information such as quality grade and flavor characteristic description.

[0021] Further, the multiple rice-producing areas in the sample preparation include Heilongjiang, Henan, Hebei, Jiangsu, Fujian, Guangdong, Jiangxi, Hubei, etc.

[0022] Further, in the physicochemical index determination and manual sensory evaluation, the sensory evaluators of the manual sensory evaluation have received strict professional training and have the ability to accurately evaluate aspects such as the aroma, taste, and appearance of rice.

[0023] Further, in the evaluation and result output, the quality grade includes excellent, good, general, and poor, and the flavor characteristic description includes aroma intensity, softness of taste, and quality of appearance color, etc.

[0024] Technical effects and advantages of the present invention:

[0025] The present invention integrates electronic nose, texture analysis and machine vision modules to accurately detect rice cooking flavor in multiple dimensions, reduce subjective influences and enhance objectivity of evaluation. Each module has a high degree of automation, and detection is efficient and fast, which can meet market demand. At the same time, it innovatively integrates multiple technologies to fill the gaps in related technologies. Its intelligent evaluation model can also be continuously optimized to adapt to diverse rice quality evaluation needs, promoting the development of food detection technology.

[0026] Specifically:

[0027] Multi-dimensional precision testing improves evaluation accuracy: The system integrates an electronic nose module, a texture analysis module, and a machine vision module to test rice cooking flavor from three key dimensions: aroma, texture, and appearance. The electronic nose module accurately detects the main aroma components of rice, the texture analysis module precisely measures multiple texture parameters to comprehensively reflect the rice's taste, and the machine vision module meticulously analyzes the rice's appearance characteristics. Multi-dimensional data fusion overcomes the limitations of traditional single evaluation methods, providing a comprehensive basis for accurately evaluating rice cooking flavor and significantly improving the accuracy and reliability of evaluation results.

[0028] Reduce subjective influence and enhance objectivity: Traditional sensory evaluation is significantly influenced by the subjective factors of the evaluator, and the evaluation results vary significantly between different evaluators. This invention greatly reduces the subjectivity of manual sensory evaluation by leveraging objective instrument detection and scientific data processing methods. The data obtained by the electronic nose, texture analyzer, and machine vision are authentic and reliable. The intelligent evaluation model constructed based on this data uses objective algorithms as the core for evaluation, making the evaluation results more fair and consistent, and more convincing in the quality comparison of rice from different batches and origins.

[0029] Efficient and rapid testing to meet market demand: Each testing module is highly automated, ensuring a rapid testing process. The electronic nose can quickly respond to rice aroma, the texture analyzer can efficiently complete simulated chewing tests, and the machine vision module rapidly collects and analyzes image data. The data processing module uses advanced algorithms to rapidly process large amounts of data, and the comprehensive evaluation module's intelligent model operates efficiently. The system can complete the testing and evaluation of a large number of samples in a short period of time, meeting the market's urgent need for rapid rice quality testing, effectively improving testing efficiency and reducing testing costs.

[0030] Integration of innovative technologies to fill technological gaps: The in - depth integration and application of optimized electronic nose technology, advanced texture analysis technology, and machine vision technology in the field of rice cooking flavor evaluation is a major innovation of this invention. Previously, electronic nose technology was less applied in the analysis of rice aroma characteristics, and machine vision technology had insufficient research in the rapid detection of the taste quality of cooked rice. This invention organically combines these technologies for the first time, develops a new intelligent sensory evaluation and detection method, fills the relevant technological gaps, provides innovative ideas and technical means for rice quality detection, and promotes the development of food detection technology.

[0031] Optimization of intelligent models to meet diverse needs: The intelligent evaluation models established by the comprehensive evaluation module, such as artificial neural network (ANN) and hierarchical clustering analysis (HCA) models, have powerful learning and optimization capabilities. During the training process, by continuously adjusting the model parameters and optimizing in combination with a large amount of artificial sensory evaluation data, the model can better adapt to the characteristics of different varieties and origins of rice, improve the accuracy and generalization ability of rice cooking flavor evaluation, and meet the diverse evaluation needs of the market for rice quality. Brief Description of the Drawings

[0032] Figure 1 It is a flowchart of an intelligent sensory evaluation and detection method for rice cooking flavor in this invention. Detailed Embodiments

[0033] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0034] It can be understood that the terms "first", "second", etc. used in this application can be used in this text to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0035] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0036] Embodiment:

[0037] Electronic nose module:

[0038] The selected electronic nose device has high sensitivity and multi-sensor characteristics, and its core lies in the construction of a targeted sensor array. In the early stage, a wealth of data was accumulated by conducting gas chromatography-mass spectrometry (GC-MS) analysis on the volatile components of a large number of different varieties of rice. Based on these data, the researchers found that substances such as aldehydes, esters, and heterocyclics are key components that affect the aroma of rice. For example, 2-acetyl-1-pyrroline, as a signature substance that gives rice a sweet smell, makes a significant contribution to its aroma. Therefore, when selecting sensors, the focus is on screening types with good response characteristics to these key aroma substances. Advanced screening technology is used to ensure that the selected sensors are not only highly sensitive to the target substances, but also have excellent selectivity, so that they can accurately capture the subtle differences in the main aroma components of rice, achieve rapid and accurate detection, and provide reliable aroma data support for subsequent flavor evaluation.

[0039] Texture Analysis Module:

[0040] Professional texture analyzers are equipped with high-precision mechanical sensors and stable mechanical transmission devices. When conducting the Texture Profile Analysis (TPA) test, standard operating procedures are strictly followed to accurately simulate the continuous compression action during oral chewing. To ensure the accuracy of the test results, the test parameters of the texture analyzer are carefully adjusted. The compression speed is set to 2 mm per second. This speed has been verified through multiple experiments and can well simulate the actual speed of oral chewing. The degree of compression is set to 60% of the original height based on the characteristics of rice to fully reflect the deformation of rice during chewing. The interval between two compressions is set to 5 seconds, allowing the rice sufficient time to recover some of its elasticity. Through these precisely set parameters, the texture parameters of rice, such as hardness, elasticity, stickiness, chewiness, and recovery, can be accurately measured, providing quantitative data for a comprehensive evaluation of the taste of rice.

[0041] Machine Vision Module:

[0042] The module consists of a high-resolution camera, professional image processing software, and a customized lighting system. The lighting system utilizes a uniform, stable, specially designed diffusely reflecting LED light source. Its illumination principle is based on advanced optical design, ensuring uniform light coverage of the sample when capturing rice images. By precisely calculating the brightness, angle, and distribution of the light source, the impact of shadows and reflections on image quality is effectively avoided. The high-resolution camera boasts 12 megapixels, enabling clear image capture of evenly detailed images of steamed rice. The professional image processing software utilizes advanced image recognition algorithms to conduct in-depth analysis of image appearance characteristics, including color, surface structure, and external shape. For example, precise extraction of CIE-L*, a*, and b* color values enables accurate assessment of rice color variations. Analysis of grain shape parameters provides insights into the plumpness and integrity of rice, providing a scientific basis for evaluating its appearance quality.

[0043] Data processing module:

[0044] Apply algorithms such as principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) to comprehensively process the collected data. In the preprocessing stage, adopt advanced filtering algorithms to remove noise interference in the electronic nose sensor signals. For example, use the adaptive filtering algorithm to automatically adjust the filtering parameters according to the real-time changes of the signals, effectively removing noises of various frequencies. Adopt the normalization method to unify the data collected by different modules to the same dimension range. For example, use the minimum-maximum normalization method to map the data to the [0,1] interval, which is convenient for subsequent feature extraction and dimensionality reduction processing, significantly improving the data processing efficiency and analysis accuracy, and laying a foundation for constructing an accurate evaluation model.

[0045] Comprehensive evaluation module:

[0046] Based on the multi-sensor fusion technology, deeply fuse the data obtained from the electronic nose, texture analysis, and machine vision. Establish intelligent evaluation models such as artificial neural network (ANN) and hierarchical cluster analysis (HCA), and conduct repeated training and optimization in combination with a large amount of artificial sensory evaluation data. When training the ANN model, continuously adjust the parameters of the model. For example, the number of layers of the neural network is set to 3 layers, the number of nodes is dynamically adjusted according to the data characteristics and model performance, and the learning rate is set to 0.001. For the HCA model, carefully select the clustering method, such as using Ward's method, and the appropriate distance metric, such as Euclidean distance, to improve the accuracy and generalization ability of the model for evaluating the cooking flavor of rice, ensuring that the model can adapt to the complex characteristics of rice from different origins and varieties, and realizing a comprehensive and accurate evaluation of the cooking flavor of rice.

[0047] Implementation steps of the detection method

[0048] Sample preparation:

[0049] A total of 80 rice samples were widely collected from multiple rice-producing areas in Heilongjiang, Henan, Hebei, Jiangsu, Fujian, Guangdong, Jiangxi, Hubei, etc. These producing areas have different geographical environments and planting conditions, and the rice produced has diverse characteristics. The rice samples were dehulled by a rice huller, which adopts advanced mechanical structures and control technologies and can efficiently remove the rice husks without damaging the rice grains. Then, the rice was milled into standard third-class precision rice in accordance with the provisions of GB1354-2009. The milling process strictly controls the milling time and intensity to ensure the consistency of rice precision. The milled rice was stored in a freezer at 2°C, which can effectively inhibit the respiration and microbial growth of the rice and maintain the freshness and quality stability of the rice. When preparing the cooked rice, it was optimized with reference to GB / T15682-2008. 10.0 g of rice samples were weighed and placed in a special aluminum box with a diameter of 45 mm and a volume of 40 ml. The material and size of the aluminum box were carefully designed to conduct heat evenly. It was washed with distilled water according to the water-to-rice ratio of 1.3:1 and soaked for 30 min in an environment at 25°C. This temperature and time can enable the rice to fully absorb water and prepare for subsequent steaming. Then it was steamed in a steamer for 40 min to ensure that the cooked rice was thoroughly cooked, and finally it was kept warm and simmered for 20 min to make the taste of the cooked rice more uniform, the starch fully gelatinized, and the quality of the cooked rice improved.

[0050] Determination of physical and chemical indexes and artificial sensory evaluation:

[0051] The gas chromatography-mass spectrometry (GC-MS) technique was used to comprehensively analyze the odors of rice of different varieties and different cooking processes. The GC-MS instrument can accurately determine the content of the main aroma substances through precise gas separation and mass spectrometry detection. For example, it can accurately detect the specific concentrations of key aroma substances such as 2-acetyl-1-pyrroline. At the same time, the quantitative descriptive method was adopted, and a group of trained professional tasters were organized to conduct artificial sensory evaluations on the rice under different cooking methods. The tasters have undergone long-term training in olfaction, taste, and vision and have sensitive perception abilities. During the evaluation process, detailed evaluations were carried out from multiple dimensions such as aroma, taste, and appearance. The richness of the aroma was described in multiple grades, and the aroma types were classified in detail, such as sweet aroma, rice aroma, etc.; the hardness, stickiness, elasticity, etc. of the taste were scored according to the standard scoring table; the color and fullness of the appearance were comprehensively evaluated, and the evaluation results were recorded in detail to construct a rich artificial sensory evaluation data set, providing an important reference basis for subsequent model training.

[0052] Multi-module detection:

[0053] The electronic nose module performs real-time detection of the aroma of rice samples based on the optimized sensor array. The sensors react with the aroma molecules volatilized from the rice, generating changes in electrical signals. These signals are converted into digital signals by a high-precision signal collector and quickly transmitted to the data processing module. The texture analysis module randomly selects 3 grains of cooked rice samples cooled to room temperature and arranges them radially on the test platform. This arrangement can more comprehensively reflect the texture uniformity of the cooked rice. Ten parallel TPA tests are carried out according to the preset instrument parameters, and various texture parameters of each test are recorded. After removing the maximum and minimum values, the average value is taken to effectively reduce the influence of abnormal data and ensure the accuracy and reliability of the data. The machine vision module acquires rice images through a high-resolution camera. The camera automatically adjusts the exposure parameters according to the light change during the acquisition process to ensure clear images. Appearance feature data such as color and shape information are extracted through image processing software analysis and transmitted to the data processing module together, providing rich visual information for comprehensive evaluation.

[0054] Data Processing and Model Construction:

[0055] The data processing module centrally processes the data transmitted by each module and uses algorithms such as PCA and PLS-DA to extract key features. The PCA algorithm can convert high-dimensional data into low-dimensional data while retaining the main features of the data, effectively reducing the data dimension and data redundancy. The PLS-DA algorithm can establish the relationship between data and categories, improving the discriminant ability of the model. The comprehensive evaluation module uses multi-sensor fusion technology to combine the processed data with artificial sensory evaluation data and constructs an intelligent sensory evaluation model using algorithms such as ANN and HCA. During the model training process, through a large number of data iterations, the model parameters are continuously optimized so that the model can accurately reflect the relationship between the cooking flavor of rice and various detection indicators, realizing the accurate prediction and evaluation of the cooking flavor of rice.

[0056] Evaluation and Result Output:

[0057] The detection data of the rice samples to be detected are input into the optimized intelligent sensory evaluation model, and the model outputs the evaluation results of the cooking flavor of rice according to the preset evaluation criteria. The evaluation criteria are formulated based on a large amount of experimental data and expert experience and are scientific and authoritative. The results include quality grades (such as excellent, good, average, poor) and flavor characteristic descriptions (such as aroma intensity, softness of taste, quality of appearance color, etc.). Through the intuitive evaluation results, it provides a scientific basis for the quality assessment, market grading of rice and consumer selection, and helps the high-quality development of the rice industry.

[0058] Each technical feature of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0059] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention should be subject to the appended claims.

[0060] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent sensory evaluation detection method for the cooking flavor of rice, characterized in that It includes an electronic nose module, a texture analysis module, a machine vision module, a data processing module, and a comprehensive evaluation module: For the electronic nose module, a high-precision and multi-sensor electronic nose device is selected. Its sensor array is optimized and screened according to the characteristics of the volatile components of rice, and includes sensors that are sensitive to the key aroma substances of rice such as aldehydes, esters, and heterocyclics, which are used to quickly and accurately detect the main aroma components of rice; For the texture analysis module, a texture analyzer with high-precision mechanical sensors and a stable mechanical transmission device is used. The TPA mode is used to conduct continuous compression tests on rice to simulate oral chewing, so as to accurately measure the texture parameters of rice such as hardness, elasticity, viscosity, chewiness, and recovery; For the machine vision module, it consists of a high-resolution camera, professional image processing software, and a lighting system. The camera is used to collect images of cooked rice, and the image processing software is used to analyze the color, surface structure, and appearance shape appearance characteristics in the image, and extract characteristic information such as CIE-L*, a*, b* color values and particle shape parameters; For the data processing module, principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) algorithms are used to preprocess, extract features, and reduce the dimension of the data collected by the electronic nose module, texture analysis module, and machine vision module; For the comprehensive evaluation module, based on multi-sensor fusion technology, the data obtained from the electronic nose, texture analysis, and machine vision are fused and processed. Intelligent evaluation models such as artificial neural network (ANN) and hierarchical cluster analysis (HCA) are established, and combined with artificial sensory evaluation data for training and optimization, so as to realize the comprehensive and objective evaluation of the cooking flavor of rice and output the evaluation results.

2. The intelligent sensory evaluation detection method for rice cooking flavor according to claim 1, characterized in that: The optimization and screening of the sensor array in the electronic nose module is based on the GC-MS analysis results of the volatile components of different varieties of rice to determine the sensor combination that is sensitive to the key aroma substances of rice.

3. The intelligent sensory evaluation detection method for rice cooking flavor according to claim 1, wherein: When the texture analyzer in the texture analysis module conducts TPA tests, the continuous compression test process that simulates oral chewing conforms to the standard operation procedure to ensure the accuracy of texture parameter measurement.

4. The intelligent sensory evaluation detection method for rice cooking flavor according to claim 1, wherein: The lighting system of the machine vision module provides uniform and stable lighting conditions to ensure the quality of the rice images collected by the high-resolution camera, which is convenient for the image processing software to accurately extract the appearance feature information.

5. A method for intelligent sensory evaluation and detection of the cooking flavor of rice according to claim 1, characterized in that: The preprocessing of the data collected by the data processing module includes noise removal and normalization operations to improve the data quality.

6. The intelligent sensory evaluation detection method for the cooking flavor of rice according to claim 1, wherein: During the training and optimization process of the intelligent evaluation model established by the comprehensive evaluation module, the model parameters are continuously adjusted to improve the accuracy and generalization ability of the model for evaluating the cooking flavor of rice.

7. An intelligent sensory evaluation and detection method for the cooking flavor of rice based on the module according to any one of claims 1-6, characterized in that, It includes the following steps: S1. Sample preparation: Collect rice samples from multiple rice-producing areas, and process them through hulling and milling to obtain standard triple-precision rice that meets the requirements of GB1354-2009. Store the rice samples in a 2°C freezer; refer to GB / T15682-2008 and improve the preparation of cooked rice. Weigh 10.0 g of rice samples into a special aluminum box, wash them with distilled water according to a water-to-rice ratio of 1.3:1, soak them at 25°C for 30 min, steam them in a steamer for 40 min, and keep them warm and simmer for 20 min; S2. Determination of physical and chemical indicators and manual sensory evaluation: Use gas chromatography-mass spectrometry (GC-MS) to analyze the odors of rice of different varieties during different cooking processes, and determine the content of the main aroma substances; adopt the quantitative description method, organize trained tasters to conduct manual sensory evaluation on rice under different cooking methods, record the evaluation results, and construct a manual sensory evaluation data set; S3. Multi-module detection: The electronic nose module detects the aroma of rice samples according to the optimized sensor array, obtains the sensor response signal and converts it into a digital signal for transmission to the data processing module; the texture analysis module conducts TPA tests on the cooked rice samples cooled to room temperature and records various texture parameters; The machine vision module collects rice images, analyzes and extracts appearance feature data through image processing software, and transmits them to the data processing module together; S4. Data processing and model construction: The data processing module processes the data transmitted by each module, and uses PCA and PLS-DA algorithms to extract features and reduce the data dimension; The comprehensive evaluation module uses multi-sensor fusion technology, combines ANN and HCA algorithms to construct an intelligent sensory evaluation model, and inputs the processed data into the model for training and optimization; S5. Evaluation and result output: Input the detection data of the rice samples to be detected into the optimized intelligent sensory evaluation model, and the model outputs the evaluation results of the cooking flavor of the rice, including the quality level and descriptive information on flavor characteristics.

8. The method for intelligent sensory evaluation of rice cooking flavor according to claim 7, wherein: The multiple rice-producing areas in the sample preparation include Heilongjiang, Henan, Hebei, Jiangsu, Fujian, Guangdong, Jiangxi, and Hubei.

9. A method for intelligent sensory evaluation and detection of the cooking flavor of rice according to claim 7, characterized in that: In the determination of physical and chemical indicators and manual sensory evaluation, the tasters for manual sensory evaluation have received strict professional training and have the ability to accurately evaluate the aroma, taste, and appearance of rice.

10. The method for intelligent sensory evaluation of rice cooking flavor according to claim 7, wherein: In the evaluation and result output, the quality level includes excellent, good, average, and poor, and the description of flavor characteristics includes aroma intensity, softness of taste, and quality of appearance color.

Images