Flower freshness evaluation system based on visual inspection

Abstract

The invention discloses a fresh flower freshness evaluation system based on visual inspection, and relates to the technical field of visual inspection, the fresh flower freshness evaluation system comprises an evaluation management center, the evaluation management center is in communication connection with an image acquisition module, an image processing module, a feature extraction module, a freshness evaluation module and an output feedback module, the modules are in electric signal connection; the image acquisition module is used for acquiring original image data of fresh flowers. According to the fresh flower freshness evaluation system based on visual inspection, a large number of fresh flowers are rapidly evaluated in a short time through the visual inspection technology, the evaluation efficiency is greatly improved, meanwhile, based on image processing and a machine learning algorithm, the system can accurately recognize the color, texture and shape features of the fresh flowers, and the evaluation accuracy is improved. Therefore, the freshness grade of the fresh flower is accurately judged, subjectivity and errors in manual evaluation are avoided, and the evaluation accuracy is improved.

Description

Technical Field

[0001] The invention relates to the technical field of visual detection, and in particular to a flower freshness assessment system based on visual detection. Background Art

[0002] Flowers are commonly used as gifts and decorations, and their freshness is extremely important to consumers. Flowers are in great demand on special occasions such as festivals, birthdays, and weddings, and consumers have increasingly higher requirements for flower quality. With the assistance of visual inspection technology, rapid, non-contact evaluation of flowers can be achieved, reducing damage and waste to flowers. For flower suppliers, freshness assessment of flowers can monitor the freshness of flowers in real time, detect and deal with unqualified flowers in a timely manner, and attract more consumers to buy, thereby improving quality control and management efficiency of flowers.

[0003] In the prior art, the freshness of flowers is affected by many factors, and the freshness of flowers is a dynamically changing process, which leads to large errors in the freshness evaluation process of flowers. Therefore, how to integrate the multimodal features of flower freshness evaluation and reduce the error of the flower freshness evaluation process is the problem we need to solve. To this end, a flower freshness evaluation system based on visual detection is proposed. Summary of the invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a flower freshness evaluation system based on visual detection, comprising an evaluation management center, wherein the evaluation management center is communicatively connected with an image acquisition module, an image processing module, a feature extraction module, a freshness evaluation module and an output feedback module, wherein the modules are electrically connected;

[0005] The image acquisition module is used to acquire the original image data of the flowers, which includes the historical image data of the flowers and the real-time image data of the flowers. The real-time image of the flowers is collected by a high-resolution camera to ensure that clear and accurate images of the flowers are captured, which provides a basis for subsequent image processing and analysis, and sets the freshness level of the flowers according to the freshness of the flowers;

[0006] The image processing module is used to pre-process the acquired raw image data of the flowers, including denoising, enhancement, and segmentation steps. The denoising is used to eliminate noise in the image and improve the image quality. The enhancement is used to improve the contrast and brightness of the image to make the features in the image more obvious. The segmentation is used to separate the flower area in the image from the background.

[0007] The feature extraction module is used to extract flower freshness features from the preprocessed flower images, where the flower freshness features include color, texture and shape, so as to analyze the overall state of the flowers;

[0008] The freshness evaluation module is used to construct a flower freshness evaluation model based on the extracted flower freshness features, evaluate the freshness of the flowers, and provide users with accurate flower freshness evaluation results;

[0009] The output feedback module is used to output the evaluation results, present them to the user in a visual manner, and collect user feedback information so that the user can intuitively understand the freshness of the flowers.

[0010] Preferably, in the image acquisition module, the process of acquiring the original image data of flowers includes:

[0011] Configure a high-resolution camera, set the resolution, focus, and exposure camera parameters of the high-resolution camera to capture clear and accurate images of flowers, and install a camera stand and light source to stabilize the camera to provide uniform lighting conditions;

[0012] The current state of flowers is captured in real time by a high-resolution camera to obtain real-time image data of flowers, ensuring that the images are clear, color-accurate, and contain the color, texture, and shape characteristics of the flowers;

[0013] The captured real-time image data of flowers is stored in the database, and each image is assigned a unique identifier for subsequent management and query;

[0014] We traverse public databases and online resources to collect historical flower image data, which contains the color, texture, and shape characteristics of the same flower with different degrees of freshness. We screen and organize the collected historical flower image data to ensure the accuracy and consistency of the data for subsequent model training.

[0015] Preferably, in the image acquisition module, the process of setting the flower freshness level includes:

[0016] The flower freshness levels are set based on the freshness of the flower historical image data, namely, the first-level flower freshness, the second-level flower freshness and the third-level flower freshness, wherein the flower freshness levels decrease from the first-level flower freshness to the third-level flower freshness;

[0017] The flower historical image data are labeled with flower freshness levels, the flower freshness level corresponding to each flower historical image is clarified, and the scoring threshold corresponding to the flower freshness level is set according to the flower freshness level and the flower freshness degree, so as to obtain a freshness level sequence table.

[0018] Preferably, in the image processing module, the preprocessing process of the original image data of the flowers includes:

[0019] Load the acquired raw image data of flowers from the database to ensure that the image data can be read and processed by the system;

[0020] Use filters to pre-process the original flower image data to eliminate noise in the image and improve the image quality and clarity. Perform image enhancement on the denoised original flower image data to improve the visual effect, contrast and brightness of the image, making the image clearer and easier to analyze.

[0021] Image segmentation is performed on the original flower image data that has been processed by denoising and image enhancement to separate the flower area from the background, and the preprocessed flower image data is saved in the database for subsequent feature extraction and classification recognition.

[0022] Preferably, the process of image segmentation of the original flower image data includes:

[0023] Use edge detection algorithms to detect edge information in images and accurately detect the edge between flowers and background;

[0024] Perform threshold processing on the edge detection results, convert the edge information into a binary image, convert the edge of the flower into the foreground, and convert the background of the flower into the background, where the foreground is white and the background is black;

[0025] Perform edge connection on the binary edge image to connect the discontinuous edges to form a complete boundary;

[0026] The edge contour of the flower is extracted from the edge-connected edge image, and the flower area in the image is separated from the background according to the edge contour to form a segmented image.

[0027] Preferably, in the feature extraction module, the process of extracting the flower freshness feature includes:

[0028] Use the image processing library to load the preprocessed flower image data, convert the image color information, convert the flower image data from RGB color space to HSV color space, and calculate the color component histogram of each color space in the image to obtain the frequency of different colors in the image to describe the color distribution of the flowers in the image;

[0029] The preprocessed flower image data is divided into small areas, the LBP algorithm is used to encode the image, the grayscale values ​​of the central pixel and the surrounding pixels are compared, and the frequencies of different encodings are counted as texture features to describe the texture features of the flowers in the image;

[0030] Use edge detection algorithm to extract edge information from preprocessed flower image data, extract contours from edge images, and calculate geometric features of contours to describe shape features of flowers in the image;

[0031] The extracted color, texture and shape features are fused to form a comprehensive feature vector, and the feature vector is normalized to facilitate subsequent classification and recognition tasks.

[0032] Preferably, in the freshness evaluation module, the construction process of the flower freshness evaluation model includes:

[0033] Extracting preprocessed historical image data of flowers with different freshness levels in the freshness level sequence table to obtain a model training data set, which is divided into a training set, a test set and a validation set;

[0034] The color, texture and shape features in the model training data set were annotated, and a linear regression model was selected to construct a flower freshness assessment model;

[0035] The data in the training set is used for model training, and the test set and validation set are used to evaluate and validate the trained flower freshness assessment model.

[0036] In the freshness evaluation module, the freshness evaluation process of flowers includes:

[0037] Applying the trained flower freshness assessment model, inputting the preprocessed flower real-time image data into the flower freshness assessment model, extracting features from the tested flower real-time image data to obtain feature vectors, and inputting the feature vectors into the flower freshness assessment model to obtain the predicted value of the flower freshness assessment;

[0038] According to the freshness evaluation prediction value of the flowers, the freshness of the flowers is classified, and the freshness level of the flowers is determined by comparing with the freshness level sequence table.

[0039] Preferably, the plurality of flower freshness levels correspond to the plurality of scoring thresholds, wherein the scoring thresholds include an upper threshold and a lower threshold;

[0040] The multiple flower freshness levels and the multiple scoring thresholds satisfy the following relationship:

[0041] First-class flower freshness R≥R y ;

[0042] Second level flower freshness R e ≤R <R y ;

[0043] Level 3 Freshness of Flowers R s ≤R <R e ;

[0044] Among them, R is the predicted value of the freshness assessment of flowers, R yis the lower threshold of the first-level flower freshness and the upper threshold of the second-level flower freshness, R e is the lower threshold of the second-level flower freshness and the upper threshold of the third-level flower freshness, R s It is the lower limit threshold of the third level of flower freshness.

[0045] Preferably, in the output feedback module, the process of collecting user feedback information includes:

[0046] According to the determined flower freshness level, the evaluation results are presented to the user on a visual interface, and different colors are used to represent different flower freshness levels. Level 1 flower freshness is represented by green, level 2 flower freshness is represented by yellow, and level 3 flower freshness is represented by red.

[0047] Display the output flower freshness level below the flower image;

[0048] Preset feedback options on the visual interface to collect user feedback on the evaluation results. Set fixed feedback options, such as "very accurate", "basically accurate", and "not very accurate", so that users can give feedback quickly.

[0049] Process and analyze the collected user feedback information, extract key information and opinions, associate the user feedback information with the evaluation results, and analyze the differences in user satisfaction and opinions at different freshness levels;

[0050] Based on user feedback, the flower freshness assessment model is optimized.

[0051] The present invention provides a flower freshness assessment system based on visual detection. It has the following beneficial effects:

[0052] 1. This flower freshness assessment system based on visual inspection uses visual inspection technology to quickly evaluate a large number of flowers in a short period of time, greatly improving the evaluation efficiency. At the same time, based on image processing and machine learning algorithms, the system can accurately identify the color, texture and shape characteristics of flowers, thereby accurately judging the freshness level of flowers, avoiding subjectivity and errors in manual evaluation and improving the accuracy of the evaluation.

[0053] 2. The flower freshness assessment system based on visual detection presents the assessment results to users in a visual manner, allowing users to intuitively understand the freshness of flowers. Displaying the freshness level of flowers with clear labels helps users quickly understand the status of flowers and facilitates users to make decisions, so as to reasonably allocate flower resources and improve user experience.

[0054] 3. The flower freshness assessment system based on visual inspection. Traditional flower freshness assessment usually needs to be performed manually, which is not only time-consuming and labor-intensive, but also easily affected by subjective factors. Flower freshness assessment based on visual inspection not only reduces labor costs, but also has higher assessment accuracy and stability, which helps save a lot of human resources and time costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a module structure diagram of a flower freshness assessment system based on visual detection according to the present invention. DETAILED DESCRIPTION

[0056] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0057] The first embodiment, as Figure 1 As shown, the present invention provides a technical solution: a flower freshness evaluation system based on visual detection, including an evaluation management center, the evaluation management center is communicatively connected with an image acquisition module, an image processing module, a feature extraction module, a freshness evaluation module and an output feedback module, wherein the modules are connected by electrical signals;

[0058] The image acquisition module is used to obtain the original image data of flowers. The original image data of flowers includes historical image data of flowers and real-time image data of flowers. Real-time images of flowers are collected by a high-resolution camera to ensure that clear and accurate images of flowers are captured, providing a basis for subsequent image processing and analysis. The historical image data of flowers are used for the training of subsequent flower freshness assessment models. The process of obtaining the original image data of flowers includes: configuring a high-resolution camera, setting the resolution, focus, and exposure camera parameters of the high-resolution camera to capture clear and accurate images of flowers, installing a camera bracket and a light source, stabilizing the camera to provide uniform lighting conditions, capturing the current state image of flowers in real time by a high-resolution camera, obtaining real-time image data of flowers, ensuring that the image is clear and color-accurate, and contains the color, texture, and shape characteristics of flowers, storing the captured real-time image data of flowers in a database, and providing Each image is assigned a unique identifier for subsequent management and query. Public databases and online resources are traversed to collect historical flower image data. The historical flower image data contains color, texture and shape features of different freshness levels of the same flower quality. The collected historical flower image data are screened and sorted to ensure the accuracy and consistency of the data for subsequent model training. The flower freshness level is set based on the freshness level of the historical flower image data, which are first-level flower freshness, second-level flower freshness and third-level flower freshness, respectively. Among them, the flower freshness level decreases step by step from the first-level flower freshness to the third-level flower freshness. The flower freshness level is annotated for the historical flower image data to clarify the flower freshness level corresponding to each historical flower image. The scoring threshold corresponding to the flower freshness level is set according to the flower freshness level and the flower freshness level, and a freshness level sequence table is obtained.

[0059] The image processing module is used to preprocess the acquired original image data of flowers, including denoising, enhancement, and segmentation steps. Denoising is used to eliminate noise in the image and improve image quality. Enhancement is used to improve the contrast and brightness of the image and make the features in the image more obvious. Segmentation is used to separate the flower area in the image from the background to facilitate subsequent feature extraction and classification recognition. The preprocessing process of the original image data of flowers includes: loading the acquired original image data of flowers from the database to ensure that the image data can be read and processed by the system, using a filter to preprocess the original image data of flowers to eliminate noise in the image and improve the quality and clarity of the image, and performing image enhancement operations on the original image data of flowers after denoising to improve the visual effect, contrast and brightness of the image and make the image clearer. The invention is easy to analyze, uses an edge detection algorithm to detect edge information in an image, accurately detects the edge between a flower and a background, performs thresholding on the edge detection result, converts the edge information into a binary image, converts the edge portion of the flower into a foreground, converts the background portion of the flower into a background, wherein the foreground is white and the background is black, performs edge connection on the binary edge image, makes the discontinuous edge connection form a complete boundary, extracts the edge contour of the flower from the edge-connected edge image, separates the flower area from the background in the image according to the edge contour, forms a segmented image, performs image segmentation on the original flower image data processed by denoising and image enhancement, separates the flower area from the background, and saves the pre-processed flower image data into a database for subsequent feature extraction and classification recognition;

[0060] The feature extraction module is used to extract flower freshness features from the preprocessed flower images. Flower freshness features include color, texture and shape, so as to analyze the overall state of the flowers. The extracted features should be able to fully reflect the freshness state of the flowers and have good discrimination and robustness. The flower freshness feature extraction process includes: using the image processing library to load the preprocessed flower image data, converting the color information of the image, converting the flower image data from the RGB color space to the HSV color space, and calculating the color component histogram of each color space in the image, obtaining the frequency of different colors in the image, so as to describe the flowers in the image. Color distribution, divide the preprocessed flower image data into small areas, use the LBP algorithm to encode the image, compare the grayscale values ​​of the central pixel and the surrounding pixels, and count the frequencies of different encodings as texture features to describe the texture features of the flowers in the image, use the edge detection algorithm to extract the edge information in the preprocessed flower image data, extract the contour from the edge image, calculate the geometric features of the contour to describe the shape characteristics of the flowers in the image, fuse the extracted color, texture and shape features to form a comprehensive feature vector, and normalize the feature vector to facilitate subsequent classification and recognition tasks.

[0061] The second embodiment is based on the first embodiment. Figure 1 As shown, the freshness evaluation module is used to construct a flower freshness evaluation model based on the extracted flower freshness features to evaluate the freshness of the flowers. The freshness evaluation module constructs and evaluates the flower freshness model based on the extracted flower freshness features to provide users with accurate flower freshness evaluation results. The flower freshness evaluation process includes: extracting pre-processed flower historical image data of different freshness levels in the freshness level sequence table to obtain a model training data set, which is divided into a training set, a test set and a validation set, annotating the color, texture and shape features in the model training data set, selecting a linear regression model to construct a freshness evaluation model, and obtaining a freshness evaluation result. A flower freshness evaluation model is built, the data in the training set is used for model training, and the trained flower freshness evaluation model is evaluated and verified by using the test set and the validation set, the trained flower freshness evaluation model is applied, the preprocessed flower real-time image data is input into the flower freshness evaluation model, feature extraction is performed on the tested flower real-time image data to obtain a feature vector, and the feature vector is input into the flower freshness evaluation model to obtain a predicted value of the flower freshness evaluation, the freshness of the flower is classified according to the predicted value of the flower freshness evaluation, and the flower freshness grade is determined by comparing with the freshness grade sequence table;

[0062] Furthermore, the multiple flower freshness levels correspond to multiple scoring thresholds, wherein the scoring thresholds include an upper threshold and a lower threshold;

[0063] Multiple flower freshness levels and multiple scoring thresholds satisfy the following relationship:

[0064] First-class flower freshness R≥R y ; R y The value is 90 points. Above 90 points represents the highest level of freshness, which usually means that the flowers are bright in color, have intact petals, are free of damage, pests and diseases, and have good water retention within a very short time after being picked.

[0065] Second level flower freshness R e ≤R <R y ; R e The value is 70 points, and 70-90 points represent medium to high freshness. The flowers may be slightly aged, but the overall condition is good, the color is bright, and the petals may be slightly damaged, but they still have high ornamental value.

[0066] Level 3 Freshness of Flowers R s ≤R <R e ; R sThe value is 60 points. 60-70 points represent a lower level of freshness. Flowers may have obvious signs of aging, such as dull color, withering petal edges, more damage or pests, but they can still be used for decoration or sold at a low price.

[0067] Among them, R is the predicted value of the freshness assessment of flowers, R y is the lower threshold of the first-level flower freshness and the upper threshold of the second-level flower freshness, R e is the lower threshold of the second-level flower freshness and the upper threshold of the third-level flower freshness, R s is the lower limit threshold of the third-level flower freshness

[0068] The output feedback module is used to output the evaluation results, present them to the user in a visual way, and collect user feedback information so that the user can intuitively understand the freshness of the flowers. The output feedback module effectively presents the flower freshness evaluation results to the user in a visual way and collects user feedback information to further optimize the evaluation system, which helps to improve user satisfaction and trust in the system and promote the continuous improvement and development of the system. The process of collecting user feedback information includes: according to the determined flower freshness level, the evaluation results are presented to the user on the visual interface, and different colors are used to represent different flower freshness levels. The first-level flower freshness is represented by green, and the second-level flower freshness is represented by blue. The first level of flower freshness is represented by yellow, and the third level of flower freshness is represented by red. The output flower freshness level is displayed below the flower image so that users can intuitively understand the freshness of the flowers. Feedback options are preset on the visual interface to collect user feedback on the evaluation results. Fixed feedback options are set, such as "very accurate", "basically accurate", "not very accurate", etc., so that users can give feedback quickly. The collected user feedback information is processed and analyzed to extract key information and opinions, and the user feedback information is associated with the evaluation results. The satisfaction and opinion differences of users at different freshness levels are analyzed, and the flower freshness evaluation model is optimized according to user feedback information.

[0069] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A flower freshness assessment system based on visual detection, comprising an assessment management center, characterized in that: The evaluation management center is communicatively connected with an image acquisition module, an image processing module, a feature extraction module, a freshness evaluation module, and an output feedback module, wherein electrical signals are connected between the modules; The image acquisition module is used to acquire the original image data of the flowers, the original image data of the flowers including the historical image data of the flowers and the real-time image data of the flowers, and to set the freshness level of the flowers according to the freshness level of the flowers; The image processing module is used to pre-process the acquired raw image data of flowers, including denoising, enhancement, and segmentation steps; The feature extraction module is used to extract flower freshness features from the preprocessed flower images, where the flower freshness features include color, texture and shape; The freshness evaluation module is used to construct a freshness evaluation model for evaluating the freshness of flowers based on the extracted freshness features of the flowers; The output feedback module is used to output the evaluation results, present them to the user in a visual manner, and collect user feedback information.

2. A flower freshness assessment system based on visual detection according to claim 1, characterized in that: In the image acquisition module, the process of acquiring the original image data of flowers includes: Configure a high-resolution camera and install the camera stand and light source; The current state image of the flowers is captured in real time by a high-resolution camera to obtain real-time image data of the flowers, which includes the color, texture and shape characteristics of the flowers; The captured real-time image data of the flowers is stored in a database and a unique identifier is assigned to each image; The historical image data of flowers are collected by traversing public databases and online resources. The historical image data of flowers contain the color, texture and shape characteristics of the same flower with different freshness levels. The collected historical image data of flowers are screened and sorted.

3. A flower freshness assessment system based on visual detection according to claim 2, characterized in that: In the image acquisition module, the process of setting the flower freshness level includes: The flower freshness levels are set based on the freshness of the flower historical image data, namely, the first-level flower freshness, the second-level flower freshness and the third-level flower freshness, wherein the flower freshness levels decrease from the first-level flower freshness to the third-level flower freshness; The flower historical image data are labeled with flower freshness levels, the flower freshness level corresponding to each flower historical image is clarified, and the scoring threshold corresponding to the flower freshness level is set according to the flower freshness level and the flower freshness degree, so as to obtain a freshness level sequence table.

4. A flower freshness assessment system based on visual detection according to claim 3, characterized in that: In the image processing module, the preprocessing process of the original image data of flowers includes: Load the obtained flower original image data from the database; Use filters to pre-process the original flower image data to eliminate noise in the image, and perform image enhancement operations on the denoised original flower image data to improve the visual effect, contrast and brightness of the image; Image segmentation is performed on the original flower image data that has been processed by denoising and image enhancement to separate the flower area from the background, and the pre-processed flower image data is saved in the database.

5. The flower freshness assessment system based on visual detection according to claim 4, characterized in that: The process of image segmentation of the flower original image data comprises: Use edge detection algorithms to detect edge information in the image and detect the edge between the flowers and the background; Perform threshold processing on the edge detection results, convert the edge information into a binary image, convert the edge of the flower into the foreground, and convert the background of the flower into the background, where the foreground is white and the background is black; Perform edge connection on the binary edge image; The edge contour of the flower is extracted from the edge-connected edge image, and the flower area in the image is separated from the background according to the edge contour to form a segmented image.

6. A flower freshness assessment system based on visual detection according to claim 5, characterized in that: In the feature extraction module, the process of extracting the flower freshness feature includes: Use the image processing library to load the preprocessed flower image data, convert the image color information, convert the flower image data from RGB color space to HSV color space, and calculate the color component histogram of each color space in the image to obtain the frequency of different colors in the image to describe the color distribution of the flowers in the image; The preprocessed flower image data is divided into small areas, the LBP algorithm is used to encode the image, the grayscale values ​​of the central pixel and the surrounding pixels are compared, and the frequencies of different encodings are counted as texture features to describe the texture features of the flowers in the image; Use edge detection algorithm to extract edge information from preprocessed flower image data, extract contours from edge images, and calculate geometric features of contours to describe shape features of flowers in the image; The extracted color, texture and shape features are fused to form a comprehensive feature vector, and the feature vector is normalized to facilitate subsequent classification and recognition tasks.

7. The flower freshness assessment system based on visual detection according to claim 6, characterized in that: In the freshness evaluation module, the construction process of the flower freshness evaluation model includes: Extracting preprocessed historical image data of flowers with different freshness levels in the freshness level sequence table to obtain a model training data set, which is divided into a training set, a test set and a validation set; The color, texture and shape features in the model training data set were annotated, and a linear regression model was selected to construct a flower freshness assessment model; The data in the training set is used for model training, and the test set and validation set are used to evaluate and validate the trained flower freshness assessment model.

8. The flower freshness assessment system based on visual detection according to claim 7, characterized in that: In the freshness evaluation module, the freshness evaluation process of flowers includes: Applying the trained flower freshness assessment model, inputting the preprocessed flower real-time image data into the flower freshness assessment model, extracting features from the tested flower real-time image data to obtain feature vectors, and inputting the feature vectors into the flower freshness assessment model to obtain the predicted value of the flower freshness assessment; According to the freshness evaluation prediction value of the flowers, the freshness of the flowers is classified, and the freshness level of the flowers is determined by comparing with the freshness level sequence table.

9. The flower freshness assessment system based on visual detection according to claim 8, characterized in that: A plurality of the flower freshness levels correspond to a plurality of the scoring thresholds, wherein the scoring thresholds include an upper threshold and a lower threshold; The multiple flower freshness levels and the multiple scoring thresholds satisfy the following relationship: First-class flower freshness R≥R y ; Second level flower freshness R e ≤R <R y ; Level 3 Freshness of Flowers R s ≤R <R e ; Among them, R is the predicted value of the freshness assessment of flowers, R y is the lower threshold of the first-level flower freshness and the upper threshold of the second-level flower freshness, R e is the lower threshold of the second-level flower freshness and the upper threshold of the third-level flower freshness, R s It is the lower limit threshold of the third level of flower freshness.

10. The flower freshness assessment system based on visual detection according to claim 9, characterized in that: In the output feedback module, the process of collecting user feedback information includes: According to the determined flower freshness level, the evaluation results are presented to the user on a visual interface, and different colors are used to represent different flower freshness levels. Level 1 flower freshness is represented by green, level 2 flower freshness is represented by yellow, and level 3 flower freshness is represented by red. Display the output flower freshness level below the flower image; Preset feedback options on the visual interface to collect user feedback on evaluation results; Process and analyze the collected user feedback information, extract key information and opinions, associate the user feedback information with the evaluation results, and analyze the differences in user satisfaction and opinions at different freshness levels; Based on user feedback, the flower freshness assessment model is optimized.

Images