Intelligent bread quality prediction and evaluation system

Abstract

The invention relates to the technical field of food production, in particular to an intelligent bread quality prediction and evaluation system which comprises a raw material monitoring module, a fermentation gas generation rate control module, a surface color and texture prediction optimization module and a quality prediction and evaluation module. The raw material monitoring module monitors characteristic data of raw materials in the bread making process in real time. The fermentation gas generation rate control module controls the generation rate of fermentation gas; the surface color and texture prediction optimization module dynamically adjusts baking parameters of an oven; the quality prediction and evaluation module predicts the final quality of the bread and generates a score. According to the invention, the expansion effect and appearance quality of the bread are always in an ideal state, the automatic adjustment of fermentation and baking parameters significantly reduces human intervention, and the automation level and quality control precision of the production process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of food production, and in particular to an intelligent bread quality prediction and evaluation system. Background Art

[0002] With the rapid development of the food industry, bread, as a daily consumer product, has increasingly higher quality requirements. In the traditional bread-making process, the stability of bread quality mainly depends on manual experience and operation, and it is difficult to maintain consistency in large-scale production. Especially in the fermentation and baking process of bread, the process parameters such as the generation rate of fermentation gas, baking temperature, and time have a decisive influence on the expansion, color and texture of the bread. How to monitor and control these key parameters in real time through intelligent means to ensure the quality consistency of bread under different production conditions is one of the important challenges facing current bread production technology.

[0003] The existing bread production process relies heavily on the experience of operators. It lacks precision and real-time adjustment capabilities in controlling key parameters such as raw material characteristics, fermentation gas generation rate, baking time and temperature, which leads to fluctuations in product quality. In addition, most existing systems are independent monitoring equipment, and there is a lack of linkage mechanism between various links, resulting in a disconnect in fermentation, baking and quality assessment, and an inability to dynamically adjust and simultaneously optimize the production process. In particular, there is a lack of intelligent monitoring methods for important appearance features such as surface color and texture, resulting in unsatisfactory baking results, low product consistency and automation levels, and the need to improve production efficiency and quality assurance.

[0004] The present invention aims to solve the above problems in the prior art, provide an intelligent prediction and evaluation system for bread quality, improve production efficiency and stability of product quality, and provide a comprehensive and intelligent control solution for the bread production process. Summary of the invention

[0005] The invention provides a bread quality intelligent prediction and evaluation system.

[0006] The intelligent bread quality prediction and evaluation system includes a raw material monitoring module, a fermentation gas generation rate control module, a surface color and texture prediction and optimization module, and a quality prediction and evaluation module, wherein; The raw material monitoring module monitors the characteristic data of raw materials in the bread making process in real time, including the protein content of flour, the purity of water, the activity of yeast and the ratio of auxiliary materials; The fermentation gas generation rate control module monitors the generation rate of carbon dioxide inside the dough in real time during the dough fermentation process, and controls the generation rate of the fermentation gas by dynamically adjusting the fermentation environment and fermentation time; The surface color and texture prediction and optimization module monitors the color and texture changes of the bread surface in real time during the bread baking process, predicts the optimal baking time and temperature of the bread, and dynamically adjusts the baking parameters of the oven, including the heating temperature and baking time, specifically including: Surface color monitoring: Real-time monitoring of the surface color changes of bread during the baking process, and comparison of the actual color data with the preset ideal color standard to analyze whether the current baking progress of the bread has reached the optimal color; Surface texture analysis: Real-time analysis of the texture characteristics of the bread surface, including the smoothness, bubble distribution, and cracks on the bread surface, and comparison with the preset standard texture data to analyze whether the current bread surface texture meets the ideal state; Dynamic adjustment of baking parameters: Based on the analysis results of surface color and surface texture, the oven baking parameters, including heating temperature and baking time, are automatically optimized; The quality prediction and evaluation module predicts the final quality of the bread and generates a score based on the monitored raw material characteristic data, the controlled fermentation gas generation rate and the adjusted oven baking parameters.

[0007] Optionally, the raw material monitoring module includes: Protein content detection: The protein content in flour is detected in real time by near infrared spectrometer (NIR), and the concentration of protein in flour is determined by the change of absorption spectrum; Water purity detection: Conduct real-time monitoring of the purity of water used for dough making through conductivity sensors, detect changes in water conductivity, determine the content of minerals and impurities in the water, and generate purity data; Yeast activity detection: The gas analyzer is used to detect the rate of carbon dioxide produced by yeast during fermentation in real time to determine the activity level of the yeast; Auxiliary material ratio monitoring: The ratio of auxiliary materials (sugar, salt, oil) is monitored in real time through the ingredient weighing equipment, and the weight of each auxiliary material is automatically recorded.

[0008] Optionally, the fermentation gas generation rate control module includes: Carbon dioxide generation rate monitoring: The integrated gas sensor monitors the carbon dioxide generation rate inside the dough in real time, captures the changes in carbon dioxide concentration during the fermentation process, and records the gas generation rate data; Fermentation environment parameter monitoring: Use temperature sensors and humidity sensors to monitor the temperature and humidity of the dough environment during the fermentation process in real time; Dynamic parameter adjustment: Based on the monitored carbon dioxide generation rate and real-time fermentation environmental parameters, the fermentation environmental parameters and fermentation time are adjusted to dynamically control the gas generation rate during the fermentation process.

[0009] Optionally, the carbon dioxide generation rate monitoring includes: Gas sensor embedding: embed the gas sensor into the fermentation container, and place the dough in the fermentation container; Carbon dioxide concentration capture: The gas sensor continuously monitors the changes in carbon dioxide concentration during the fermentation process, and identifies the amount of gas generated and the progress of fermentation by reacting to the absorption and release of carbon dioxide; Generation rate data recording: Based on the changes in carbon dioxide concentration captured by the gas sensor, the carbon dioxide generation rate is automatically calculated and recorded in the database.

[0010] Optionally, the dynamic parameter adjustment includes: Calculate the deviation from the target value: The real-time monitored carbon dioxide generation rate With the preset target rate Compare and calculate deviation ; Adjust temperature parameters: dynamically adjust the temperature in the fermentation environment according to the deviation of the carbon dioxide production rate; Adjust humidity parameters: Automatically adjust the humidity of the fermentation environment according to the deviation of the monitored carbon dioxide generation rate to ensure that the humidity is within the ideal range; Fermentation time adjustment: Automatically adjust the fermentation time according to the changing trend of the carbon dioxide generation rate.

[0011] Optionally, the surface color monitoring includes: Multispectral imaging data acquisition: Through the integrated multispectral camera, the light reflection data of the bread surface under different spectra is captured in real time, and the surface color change information of the bread under different wavelengths during the baking process is obtained; Color data conversion and extraction: Convert light reflectance data obtained by multispectral imaging into brightness in color space through XYZ color space , red and green axis and blue and yellow axis The value of is used to analyze the surface color of bread; Color difference calculation: Through the color difference analysis algorithm, the actual color data monitored is compared with the preset ideal color standard to calculate the color difference ; Baking progress analysis: based on the calculated color difference , compared with the set color difference threshold. If the color difference is less than the color difference threshold, it means that the bread surface has reached the ideal color.

[0012] Optionally, the surface texture analysis includes: Image data acquisition: The image data of the bread surface is collected in real time through the camera; Texture feature extraction: Use a convolutional neural network (CNN) model to extract texture features from image data, including the smoothness, bubbles, and cracks on the bread surface; Compare with standard texture data: Compare the extracted actual texture feature data with the preset standard texture data and calculate the deviation , analyze whether the texture state of the current bread surface meets the ideal state; Progress Analysis: Based on Deviation Determine whether the bread baking progress is up to standard. Exceeding the preset texture threshold indicates that the bread baking progress is not up to standard.

[0013] Optionally, the convolutional neural network (CNN) model includes: Multi-scale convolutional layer design: A multi-scale convolutional layer is used to capture texture features at different scales (such as large bubbles, small cracks, etc.). and ), extracting large-scale and small-scale texture feature information at the same time; Texture-aware pooling layer: Use the texture-aware pooling layer to extract the texture information of the bread surface and adaptively adjust the pooling window according to the changes in local texture features; Multi-task loss function: Use the multi-task loss function to process the texture features of smoothness, bubbles, and cracks at the same time, calculate the loss of each feature separately, and perform weighted summation; Output layer: In the final output layer, the characteristic values ​​of smoothness, bubbles and cracks are output respectively.

[0014] Optionally, the dynamic adjustment of baking parameters includes: Surface color analysis: By monitoring the surface color changes during bread baking, the color difference is calculated to determine whether the current color has reached the optimal color; Surface texture analysis: By analyzing the texture characteristics of the bread surface (smoothness, bubble distribution, cracks) in real time, the texture deviation is calculated to determine whether the texture state of the bread surface meets the ideal state; Parameter optimization calculation: Based on the color and texture analysis results, the proportional integral derivative (PID) control algorithm is used to dynamically adjust the oven's heating temperature and baking time.

[0015] Optionally, the quality prediction and evaluation module includes: Raw material characteristic score: The raw material characteristic score is generated by weighting the protein content of flour, water purity, yeast activity and the ratio of auxiliary materials. ; Fermentation score: Calculate the fermentation score based on the adjusted fermentation gas production rate ; Roasting score: Calculates the roasting score based on the adjusted roasting temperature and time ; Final quality score: The final quality score is calculated by weighting the fermentation score, baking score and raw material score. .

[0016] Beneficial effects of the present invention: The present invention, through an intelligent raw material monitoring module, can monitor in real time the key characteristic data of raw materials in the bread-making process, including the protein content of flour, the purity of water, the activity of yeast and the ratio of auxiliary materials. Through precise raw material detection, the system ensures the quality stability and consistency of each batch of bread-making raw materials, reduces the difference in finished products caused by fluctuations in raw material quality, and thus provides reliable data support for the fermentation and baking processes. This precise raw material monitoring not only effectively avoids human errors, but also improves overall production efficiency.

[0017] The present invention can realize real-time monitoring and dynamic adjustment of bread fermentation and baking processes through the fermentation gas generation rate control module and the surface color and texture prediction and optimization module. By accurately controlling the carbon dioxide generation rate, fermentation environment parameters and color and texture in the fermentation process, the system can effectively optimize the fermentation and baking parameters to ensure that the expansion effect and appearance quality of the bread are always in an ideal state. The automatic adjustment of the fermentation and baking parameters significantly reduces human intervention and improves the automation level and quality control accuracy of the production process.

[0018] The present invention combines a comprehensive scoring system of raw material characteristics, fermentation process and baking parameters through a quality prediction and evaluation module, and can dynamically evaluate and predict the final quality of bread. By weighted calculation of raw material scores, fermentation scores and baking scores, the system can accurately predict core quality characteristics of bread such as appearance, texture and flavor, thereby ensuring the stability and high-quality output of bread production, improving the accuracy and intelligence level of the entire production process, making each link in the production interrelated and mutually supportive, and ensuring the consistency and high quality of the final product. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention; Figure 2Schematic diagram of a surface color and texture prediction optimization module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0022] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0023] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0024] like Figure 1-Figure 2 As shown, the bread quality intelligent prediction and evaluation system includes a raw material monitoring module, a fermentation gas generation rate control module, a surface color and texture prediction and optimization module, and a quality prediction and evaluation module, wherein; The raw material monitoring module monitors the characteristic data of raw materials in the bread-making process in real time, including the protein content of flour, water purity, yeast activity and the ratio of auxiliary materials; The fermentation gas generation rate control module monitors the generation rate of carbon dioxide in the dough in real time during the dough fermentation process, and controls the generation rate of fermentation gas by dynamically adjusting the fermentation environment and fermentation time, ensuring that the gas generation rate is stable within the ideal range and optimizing the expansion effect of the bread; The surface color and texture prediction and optimization module monitors the color and texture changes of the bread surface in real time during the bread baking process, predicts the optimal baking time and temperature of the bread, and dynamically adjusts the baking parameters of the oven, including heating temperature and baking time, to ensure uniform color and ideal texture of the bread surface. Specifically, it includes: Surface color monitoring: Real-time monitoring of the surface color changes of bread during the baking process, and comparison of the actual color data with the preset ideal color standard to analyze whether the current baking progress of the bread has reached the optimal color; Surface texture analysis: Real-time analysis of the texture characteristics of the bread surface, including the smoothness, bubble distribution, and cracks on the bread surface, and comparison with the preset standard texture data to analyze whether the current bread surface texture meets the ideal state; Dynamic adjustment of baking parameters: Based on the analysis results of surface color and surface texture, the oven baking parameters, including heating temperature and baking time, are automatically optimized; The quality prediction and evaluation module predicts the final quality of the bread and generates a score based on the monitored raw material characteristic data, the controlled fermentation gas generation rate and the adjusted oven baking parameters; Through the above content, intelligent management of the entire process is achieved, from raw material detection, fermentation gas generation rate control, color and texture monitoring during baking to quality evaluation. It can dynamically adjust fermentation and baking parameters to ensure the quality consistency and optimized control of bread under different production conditions. It not only improves production efficiency, but also significantly improves the stability of the appearance, texture and flavor of bread, reduces human intervention and quality fluctuations, and provides a more accurate and automated solution for bread production.

[0025] The raw material monitoring module includes: Protein content detection: The protein content in flour is detected in real time by near infrared spectrometer (NIR), and the concentration of protein in flour is determined by the change of absorption spectrum; Water purity detection: Conduct real-time monitoring of the purity of water used for dough making through conductivity sensors, detect changes in water conductivity, determine the content of minerals and impurities in the water, and generate purity data; Yeast activity detection: The gas analyzer is used to detect the rate of carbon dioxide produced by yeast during fermentation in real time to determine the activity level of the yeast; Auxiliary material ratio monitoring: The ratio of auxiliary materials (sugar, salt, oil) is monitored in real time through the ingredient weighing equipment, and the weight of each auxiliary material is automatically recorded; Through the above content, it is ensured that the quality and characteristics of each raw material in the bread-making process remain within the ideal range, which can reduce the impact of raw material quality fluctuations on bread production, improve production consistency and stability, avoid human operation errors, and provide reliable basic data support for subsequent fermentation and baking processes. Through precise raw material monitoring, the prerequisites for bread quality are guaranteed to ensure that the flavor and texture of the final product meet the expected standards.

[0026] The fermentation gas generation rate control module includes: Carbon dioxide generation rate monitoring: The integrated gas sensor monitors the carbon dioxide generation rate inside the dough in real time, captures the changes in carbon dioxide concentration during the fermentation process, and records the gas generation rate data; Fermentation environment parameter monitoring: The temperature and humidity of the dough environment during the fermentation process are monitored in real time through temperature sensors and humidity sensors to ensure that the fermentation environment is within the ideal range; Dynamic parameter adjustment: Based on the monitored carbon dioxide generation rate and real-time fermentation environment parameters, the fermentation environment parameters and fermentation time are adjusted to dynamically control the gas generation rate during the fermentation process; Through the above content, the gas generation and environmental conditions during the fermentation process are precisely controlled, the expansion effect of the dough is effectively optimized, and insufficient or excessive fermentation is avoided. Through automated monitoring and adjustment, the stability and consistency of the fermentation process are improved, which greatly reduces the need for human intervention and ensures the quality consistency and controllability of bread in different batches.

[0027] Carbon dioxide generation rate monitoring includes: Gas sensor embedding: embed the gas sensor into the fermentation container, and place the dough in the fermentation container; Carbon dioxide concentration capture: The gas sensor continuously monitors the changes in carbon dioxide concentration during the fermentation process, and identifies the amount of gas generated and the progress of fermentation by reacting to the absorption and release of carbon dioxide; Generation rate data recording: According to the changes in carbon dioxide concentration captured by the gas sensor, the carbon dioxide generation rate is automatically calculated and recorded in the database, expressed as: ; in, is the rate of carbon dioxide production, represents the change in carbon dioxide concentration, Indicates a time interval; Through the above content, the fermentation status can be reflected in real time and accurately. By monitoring the gas generation rate, the system can adjust the fermentation environment in time, optimize the fermentation effect, and avoid excessive or insufficient fermentation. This monitoring method not only improves the control accuracy of the fermentation process, but also can greatly reduce human intervention and ensure the consistency and quality stability of bread production.

[0028] Dynamic parameter adjustments include: Calculate the deviation from the target value: The real-time monitored carbon dioxide generation rate With the preset target rate Compare and calculate deviation , expressed as: ; in, is the deviation of the carbon dioxide generation rate; Adjust temperature parameters: According to the deviation of carbon dioxide production rate, dynamically adjust the temperature in the fermentation environment to ensure that the fermentation temperature is optimized to the ideal range, expressed as: ; in, is the adjusted temperature, is the current temperature, is the temperature regulation coefficient, is the deviation of the carbon dioxide generation rate; Adjust humidity parameters: According to the deviation of the monitored carbon dioxide generation rate, the humidity of the fermentation environment is automatically adjusted to ensure that the humidity is within the ideal range, expressed as: ; in, is the adjusted humidity, is the current humidity, For humidity regulation; Fermentation time adjustment: According to the changing trend of carbon dioxide generation rate, the fermentation time is automatically adjusted to ensure that the dough ferments under ideal conditions, expressed as: ; in, is the adjusted fermentation time, is the current fermentation time, is the target carbon dioxide generation rate, is the real-time monitored carbon dioxide generation rate. is the fermentation time adjustment coefficient, which is used to control the sensitivity of time adjustment; Through the above content, the fermentation process can be accurately controlled to ensure that the carbon dioxide generation rate always remains within the ideal range, avoiding insufficient or excessive fermentation. By adjusting the fermentation time, it is also possible to flexibly respond to environmental fluctuations and optimize the dough expansion effect. This automated adjustment method improves the stability and consistency of the fermentation process, greatly reduces human intervention, and ensures stable bread quality and higher production efficiency.

[0029] Surface color monitoring includes: Multispectral imaging data acquisition: Through the integrated multispectral camera, the light reflection data of the bread surface under different spectra is captured in real time, and the surface color change information of the bread under different wavelengths during the baking process is obtained; Color data conversion and extraction: Convert light reflectance data obtained by multispectral imaging into brightness in color space through XYZ color space , red and green axis and blue and yellow axis The value of is used to analyze the surface color of bread; Color difference calculation: Through the color difference analysis algorithm, the actual color data monitored is compared with the preset ideal color standard to calculate the color difference , expressed as: ; in, They are the currently monitored color brightness and color axis value, is the preset ideal color standard value, is the color difference between the actual color and the ideal color; Baking progress analysis: based on the calculated color difference , compared with the set color difference threshold. If the color difference is less than the color difference threshold, it means that the bread surface has reached the ideal color; XYZ color space specifically includes: (1) Conversion from light reflection data to XYZ color space: The multispectral data is converted to the CIE 1931 XYZ color space. The XYZ color space is obtained by weighted averaging the light reflection data and is expressed as: ; ; ; in, is the spectral distribution of the light source, is the chromaticity function value of the CIE 1931 standard observer, is the reflectivity of the object surface to a specific wavelength; (2) Convert from XYZ color space to Lab color space: Convert the value of XYZ color space to CIE Lab color space, expressed as: ; ; ; in, is the XYZ value of the reference white point; ; The color difference threshold is set based on historical color data, including: Collect historical data: record the color difference of each batch of bread during multiple baking value; Analyze ideal baking color data: extract baking batches with high quality evaluation and color close to the ideal standard from historical data, and calculate the average and standard deviation of the color difference values ​​of these batches according to their color difference values; Set threshold: Set the threshold based on the color difference value range of these ideal batches, usually the average value Plus a certain standard deviation As the upper threshold of color difference, ensure that the baking parameters are adjusted only when the color difference value is too large, expressed as: ; in, is the color difference threshold set, is the average color difference of the ideal baking batch in the historical data, is the color difference standard deviation of the ideal baking batch in the historical data, is the adjustment factor, Between, used to control sensitivity to color fluctuations; Through the above content, we can ensure that the bread reaches the best color during the baking process. We can automatically judge the baking progress according to the preset color difference threshold, adjust the baking parameters in time to avoid uneven color or over-baking. The color difference threshold set based on historical data makes monitoring more accurate and reliable, effectively improving the consistency of baking quality and the controllability of product appearance.

[0030] Surface texture analysis includes: Image data acquisition: The image data of the bread surface is collected in real time through the camera; Texture feature extraction: Use a convolutional neural network (CNN) model to extract texture features from image data, including the smoothness, bubbles, and cracks on the bread surface; Compare with standard texture data: Compare the extracted actual texture feature data with the preset standard texture data and calculate the deviation , analyze whether the texture state of the current bread surface meets the ideal state, expressed as: ; in, is the deviation between the actual texture feature data and the standard texture data, is the current surface smoothness feature extracted from real-time image data, is the current bubble distribution feature extracted from real-time image data, is the current crack density feature extracted from the real-time image data, It is the preset ideal texture standard value; Progress Analysis: Based on Deviation Determine whether the bread baking progress is up to standard. Exceeding the preset texture threshold indicates that the bread baking progress is not up to standard; The texture threshold is set based on historical image data and is expressed as: Collect historical data: record the baking process of multiple batches of bread and obtain the texture feature data of each batch (smoothness, bubble distribution, crack density, etc.); Extract the deviation value of texture features: Calculate the deviation of texture features for each batch ; Analyze the deviations from the ideal roasted batch: Extract the best roasted batches from the historical data and calculate the deviations from these batches and standard deviation ; Set the threshold: It is determined by adding a certain standard deviation to the mean deviation value of the ideal batch to cover the normal deviation range, expressed as: ; in, is the texture threshold, is the mean of the ideal batch deviation values, is the standard deviation of the ideal batch deviation values, is the adjustment factor (between 1 and 2) used to adjust the sensitivity of the threshold; Through the above content, the surface quality of bread during the baking process can be accurately evaluated. Combined with the convolutional neural network model, texture features can be automatically extracted to reduce human detection errors and ensure that the texture and appearance of the bread surface meet the preset standards. The thresholds set based on historical data can also further optimize automatic control, adjust baking parameters in time, improve production efficiency and product consistency, and ensure the high quality of the final bread. This precise texture analysis improves the ability to control product quality and makes the entire baking process more intelligent and reliable.

[0031] The Convolutional Neural Network (CNN) model consists of: Multi-scale convolutional layer design: A multi-scale convolutional layer is used to capture texture features at different scales (such as large bubbles, small cracks, etc.). and ), and extract large-scale and small-scale texture feature information at the same time, expressed as: ; in, It is Output feature map after layer convolution, is the weight matrix of the convolution kernel, is the pixel value of the input feature map (output of the previous layer), It is The bias of the layer, , is the size of the convolution kernel (e.g. and wait); Texture-aware pooling layer: Use the texture-aware pooling layer to extract the texture information of the bread surface and adaptively adjust the pooling window according to the changes in local texture features, expressed as: ; in, For the The feature value after layer pooling, is the convolution feature map with Neighboring pixel values ​​(adaptively adjusted based on the pooling window), is the position offset in the adaptive pooling window, To adaptively size the pooling region, the range of the window is dynamically adjusted based on the complexity of local texture features; Multi-task loss function: Use the multi-task loss function to process different texture features such as smoothness, bubbles, and cracks at the same time, calculate the loss of each feature separately, and perform weighted summation, which is expressed as: ; in, is the total loss function, and are the loss of smoothness, bubbles and cracks, is a weight parameter used to balance the losses between different tasks; ; in, is the number of pixel pairs involved in the calculation in the image, For location The pixel value of For location The adjacent pixel values ​​of ; in, is the total number of bubbles detected, For the The actual area of ​​the bubble, For the The ideal standard area of ​​a bubble; ; in, is the total number of cracks detected, For the The actual length of the crack, For the The ideal standard length of a crack; Output layer: In the final output layer, the characteristic values ​​of smoothness, bubbles and cracks are output respectively, expressed as: ; ; ; in, is the smoothness eigenvalue of the output, is the output bubble distribution characteristic value, is the output crack density eigenvalue, , , are the weight matrices of the fully connected layer for smoothness, bubble distribution, and crack density, respectively. is the feature map of the last convolutional layer, To flatten the last layer of convolutional feature maps into a one-dimensional vector, , , are the bias terms of the fully connected layer for smoothness, bubble distribution, and crack density respectively; Through the above content, different texture features of the bread surface, such as smoothness, bubble distribution and crack density, can be efficiently extracted, and texture information of different scales and complexities can be processed simultaneously, thereby improving the detection accuracy of surface quality. By using the fully connected layer to output specific texture feature values, the model can provide real-time data support for subsequent quality evaluation and baking parameter adjustment, thereby improving the intelligence and automation level of the bread production process and ensuring product consistency and high quality.

[0032] Dynamic adjustment of baking parameters includes: Surface color analysis: By monitoring the surface color changes during bread baking, the color difference is calculated to determine whether the current color has reached the optimal color; Surface texture analysis: By analyzing the texture characteristics of the bread surface (smoothness, bubble distribution, cracks) in real time, the texture deviation is calculated to determine whether the texture state of the bread surface meets the ideal state; Parameter optimization calculation: Based on the color and texture analysis results, the proportional integral derivative (PID) algorithm is used to dynamically adjust the oven's heating temperature and baking time to ensure that the bread's surface color and texture are within the ideal range, expressed as: ; in, is the adjusted baking temperature, is the initial baking temperature. is the deviation value of the surface texture, is the color difference deviation value, are the proportional, integral and differential coefficients in the PID control algorithm; Through the above content, the color difference and texture deviation of the bread surface can be comprehensively considered, and the heating temperature and baking time of the oven can be automatically optimized to ensure that the bread achieves the ideal baking effect. The introduction of color difference and texture deviation calculation makes the adjustment more precise, and can respond to subtle changes in the baking process in time to avoid excessive or insufficient baking. This dynamic adjustment based on the PID control algorithm improves the consistency of baking quality, reduces the need for human intervention, and ensures the intelligence and efficiency of the production process.

[0033] The quality prediction and assessment modules include: Raw material characteristic score: The raw material characteristic score is generated by weighting the protein content of flour, water purity, yeast activity and the ratio of auxiliary materials. , expressed as: ; in, is the protein content, For the purity of water, For the activity of yeast, For the auxiliary material ratio, is the weight of each raw material characteristic, Score raw materials; Fermentation score: Calculate the fermentation score based on the adjusted fermentation gas production rate , expressed as: ; in, is the adjusted carbon dioxide generation rate, is the ideal carbon dioxide generation rate, Score the fermentation; Roasting score: Calculates the roasting score based on the adjusted roasting temperature and time , expressed as: ; in, is the adjusted baking temperature, For the ideal baking temperature, For the adjusted baking time, For the ideal baking time, is the weight coefficient of baking temperature and time, satisfying ; Final quality score: The final quality score is calculated by weighting the fermentation score, baking score and raw material score. , expressed as: ; in, Score the final quality. is the weight of each score, indicating the impact of each score on the final quality score; Through the above content, real-time evaluation can be dynamically performed based on the adjusted actual process parameters to ensure that the fermentation and baking processes at each stage are within the ideal control range. By weighted calculation of the fermentation score, baking score and raw material score, the accuracy of the prediction is effectively improved, human intervention is reduced, and a comprehensive evaluation and optimized control of bread quality is achieved, thereby ensuring product consistency and high quality.

[0034] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0035] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. Bread quality intelligent prediction and evaluation system, characterized by: It includes a raw material monitoring module, a fermentation gas generation rate control module, a surface color and texture prediction and optimization module, and a quality prediction and evaluation module, among which; The raw material monitoring module monitors the characteristic data of raw materials in the bread making process in real time, including the protein content of flour, the purity of water, the activity of yeast and the ratio of auxiliary materials; The fermentation gas generation rate control module monitors the generation rate of carbon dioxide inside the dough in real time during the dough fermentation process, and controls the generation rate of the fermentation gas by dynamically adjusting the fermentation environment and fermentation time; The surface color and texture prediction and optimization module monitors the color and texture changes of the bread surface in real time during the bread baking process, predicts the optimal baking time and temperature of the bread, and dynamically adjusts the baking parameters of the oven, including the heating temperature and baking time, specifically including: Surface color monitoring: Real-time monitoring of the surface color changes of bread during the baking process, and comparison of the actual color data with the preset ideal color standard to analyze whether the current baking progress of the bread has reached the optimal color; Surface texture analysis: Real-time analysis of the texture characteristics of the bread surface, including the smoothness, bubble distribution, and cracks on the bread surface, and comparison with the preset standard texture data to analyze whether the current bread surface texture meets the ideal state; Dynamic adjustment of baking parameters: Based on the analysis results of surface color and surface texture, the oven baking parameters, including heating temperature and baking time, are automatically optimized; The quality prediction and evaluation module predicts the final quality of the bread and generates a score based on the monitored raw material characteristic data, the controlled fermentation gas generation rate and the adjusted oven baking parameters.

2. The bread quality intelligent prediction and evaluation system according to claim 1, characterized in that: The raw material monitoring module comprises: Protein content detection: The protein content in flour is detected in real time by near-infrared spectrometer, and the concentration of protein in flour is determined by the change of absorption spectrum; Water purity detection: Conduct real-time monitoring of the purity of water used for dough making through conductivity sensors, detect changes in water conductivity, determine the content of minerals and impurities in the water, and generate purity data; Yeast activity detection: The gas analyzer is used to detect the rate of carbon dioxide produced by yeast during fermentation in real time to determine the activity level of the yeast; Auxiliary material ratio monitoring: The auxiliary material ratio is monitored in real time through the ingredient weighing equipment, and the weight of each auxiliary material is automatically recorded.

3. The bread quality intelligent prediction and evaluation system according to claim 1, characterized in that: The fermentation gas generation rate control module comprises: Carbon dioxide generation rate monitoring: The integrated gas sensor monitors the carbon dioxide generation rate inside the dough in real time, captures the changes in carbon dioxide concentration during the fermentation process, and records the gas generation rate data; Fermentation environment parameter monitoring: Use temperature sensors and humidity sensors to monitor the temperature and humidity of the dough environment during the fermentation process in real time; Dynamic parameter adjustment: Based on the monitored carbon dioxide generation rate and real-time fermentation environmental parameters, the fermentation environmental parameters and fermentation time are adjusted to dynamically control the gas generation rate during the fermentation process.

4. The bread quality intelligent prediction and evaluation system according to claim 3, characterized in that: The carbon dioxide generation rate monitoring includes: Gas sensor embedding: embed the gas sensor into the fermentation container, and place the dough in the fermentation container; Carbon dioxide concentration capture: The gas sensor continuously monitors the changes in carbon dioxide concentration during the fermentation process, and identifies the amount of gas generated and the progress of fermentation by reacting to the absorption and release of carbon dioxide; Generation rate data recording: Based on the changes in carbon dioxide concentration captured by the gas sensor, the carbon dioxide generation rate is automatically calculated and recorded in the database.

5. The bread quality intelligent prediction and evaluation system according to claim 4, characterized in that: The dynamic parameter adjustment includes: Calculate the deviation from the target value: The real-time monitored carbon dioxide generation rate With the preset target rate Compare and calculate deviation ; Adjust temperature parameters: dynamically adjust the temperature in the fermentation environment according to the deviation of the carbon dioxide production rate; Adjust humidity parameters: Automatically adjust the humidity of the fermentation environment according to the deviation of the monitored carbon dioxide generation rate to ensure that the humidity is within the ideal range; Fermentation time adjustment: Automatically adjust the fermentation time according to the changing trend of the carbon dioxide generation rate.

6. The bread quality intelligent prediction and evaluation system according to claim 1, characterized in that: The surface color monitoring includes: Multispectral imaging data acquisition: Through the integrated multispectral camera, the light reflection data of the bread surface under different spectra is captured in real time, and the surface color change information of the bread under different wavelengths during the baking process is obtained; Color data conversion and extraction: Convert the light reflectance data obtained by multispectral imaging into brightness in color space through XYZ color space , red and green axis and blue and yellow axis The value of is used to analyze the surface color of bread; Color difference calculation: Through the color difference analysis algorithm, the actual color data monitored is compared with the preset ideal color standard to calculate the color difference ; Baking progress analysis: based on the calculated color difference , compared with the set color difference threshold. If the color difference is less than the color difference threshold, it means that the bread surface has reached the ideal color.

7. The bread quality intelligent prediction and evaluation system according to claim 6, characterized in that: The surface texture analysis includes: Image data acquisition: The image data of the bread surface is collected in real time through the camera; Texture feature extraction: Use a convolutional neural network model to extract texture features from image data, including the smoothness, bubbles, and cracks on the bread surface; Compare with standard texture data: Compare the extracted actual texture feature data with the preset standard texture data and calculate the deviation , analyze whether the texture state of the current bread surface meets the ideal state; Progress Analysis: Based on Deviation Determine whether the bread baking progress is up to standard. Exceeding the preset texture threshold indicates that the bread baking progress is not up to standard.

8. The bread quality intelligent prediction and evaluation system according to claim 7, characterized in that: The convolutional neural network model includes: Multi-scale convolution layer design: A multi-scale convolution layer is used to capture texture features at different scales. By using convolution kernels of different sizes, both large-scale and small-scale texture feature information can be extracted simultaneously. Texture-aware pooling layer: Use the texture-aware pooling layer to extract the texture information of the bread surface and adaptively adjust the pooling window according to the changes in local texture features; Multi-task loss function: Use the multi-task loss function to process the texture features of smoothness, bubbles, and cracks at the same time, calculate the loss of each feature separately, and perform weighted summation; Output layer: In the final output layer, the characteristic values ​​of smoothness, bubbles and cracks are output respectively.

9. The bread quality intelligent prediction and evaluation system according to claim 8, characterized in that: The dynamic adjustment of baking parameters includes: Surface color analysis: By monitoring the surface color changes during bread baking, the color difference is calculated to determine whether the current color has reached the optimal color; Surface texture analysis: By analyzing the texture characteristics of the bread surface in real time, the texture deviation is calculated to determine whether the texture state of the bread surface meets the ideal state; Parameter optimization calculation: Based on the color and texture analysis results, the proportional integral derivative control algorithm is used to dynamically adjust the oven's heating temperature and baking time.

10. The bread quality intelligent prediction and evaluation system according to claim 1, characterized in that: The quality prediction and evaluation module includes: Raw material characteristic score: The raw material characteristic score is generated by weighting the protein content of flour, water purity, yeast activity and the ratio of auxiliary materials. ; Fermentation score: Calculate the fermentation score based on the adjusted fermentation gas production rate ; Roasting score: Calculates the roasting score based on the adjusted roasting temperature and time ; Final quality score: The final quality score is calculated by weighting the fermentation score, baking score and raw material score. .

Images