Intelligent image recognition method and system based on automatic cooker

Abstract

The invention relates to an intelligent image recognition method and system based on an automatic cooker, and belongs to the technical field of image recognition. The method comprises the following steps: acquiring food material image data through a built-in visual sensor of a feeding bin of the automatic cooker, and calculating through a preprocessing model according to the food material image data to obtain image preprocessing data; according to the image preprocessing data, a food material recognition result sequence is obtained through calculation of a recognition model; and traversing a database according to the food material identification result sequence to obtain a cooking recipe, and transmitting the cooking recipe to a human-computer interaction interface by the intelligent terminal. Automatic identification of food materials and automatic menu display are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of image recognition, and in particular relates to an intelligent image recognition method and system based on an automatic cooking machine. Background Art

[0002] With the rapid development of modern technology, smart home devices have gradually become an important part of people's daily lives. Among them, automatic cooking machines, as kitchen appliances that combine convenience, efficiency, and intelligence, are gaining increasing popularity in the market. However, most automatic cooking machines currently on the market rely on preset cooking programs or manual input of ingredient information to perform cooking tasks. They lack the ability to automatically identify ingredients and flexibly respond to them, which to some extent limits their intelligence and the optimization of user experience.

[0003] The application of traditional image recognition technology in automatic cooking machines faces numerous challenges. For one thing, factors such as fluctuating lighting, diverse ingredient shapes, and complex backgrounds in the kitchen environment make image preprocessing a critical and complex task. While traditional preprocessing methods such as normalization, grayscale conversion, and noise reduction can improve image quality to a certain extent, they often struggle to effectively cope with complex and ever-changing kitchen scenes, resulting in low subsequent recognition accuracy. Furthermore, existing image recognition models often suffer from slow recognition speeds and low accuracy when handling ingredient recognition, especially when faced with complex situations such as mixed ingredients, partial occlusion, or morphological variations. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides an intelligent image recognition method and system based on an automatic cooking machine. The purpose of the present invention can be achieved through the following technical solutions: The food image data is acquired by a built-in visual sensor in the feed bin of the automatic cooking machine, and image preprocessing data is calculated based on the food image data by a preprocessing model; Obtaining a food identification result sequence by calculating the image preprocessing data through a recognition model; The database is traversed according to the ingredient recognition result sequence to obtain a cooking recipe, and the intelligent terminal transmits the cooking recipe to the human-computer interaction interface.

[0005] Specifically, the preprocessing model calculation step includes: Normalizing the food image data using a linear function conversion method to obtain normalized data; Gray-scaling the normalized data using a weighted average method to obtain gray-scale data; Performing noise reduction processing on the grayscale data by median filtering to obtain noise-reduced data; The image preprocessing data is obtained by calculating a reconstruction model according to the noise reduction data.

[0006] Specifically, the reconstruction model includes a shallow feature extraction subunit, a deep feature extraction subunit, and a feature fusion subunit. The specific calculation steps include: Obtain shallow features by calculating the shallow feature extraction subunit according to the noise reduction data; A deep feature extraction subunit calculates deep sub-features based on the denoised data, and performs feature fusion using a 1*1 convolution kernel based on the deep sub-features to obtain deep features; The image preprocessing data is obtained by calculation through a feature fusion subunit according to the shallow features and the deep features.

[0007] Specifically, the recognition model calculation step includes: Perform image segmentation using a U-net neural network based on the image preprocessing data to obtain an image of the area to be identified; Obtaining a recognition result by calculating the image of the area to be recognized through the VGGNet model; Each of the identification results is stored to obtain the food identification result sequence.

[0008] Specifically, traversing the database according to the ingredient identification result sequence to obtain a cooking recipe specifically includes: S201: Preset a traversal pointer, and sequentially traverse the food identification result sequence according to the traversal pointer to obtain a search identification result; S202: Traversing the database according to the search and identification results to obtain a recipe subsequence; S203: Determine whether the traversal pointer is empty. If yes, execute step S204; if no, repeat steps S201-S203; S204: Obtain the cooking recipe through a union-find calculation based on the recipe subsequence.

[0009] An intelligent image recognition system based on an automatic cooking machine includes: an acquisition unit, a recognition unit, and a query unit; The acquisition unit is used to obtain food image data through a built-in visual sensor in the feed bin of the automatic cooking machine, and obtain image preprocessing data based on the food image data through calculation using a preprocessing model; The recognition unit is configured to calculate a food identification result sequence based on the image preprocessing data through a recognition model; The query unit is used to traverse the database according to the ingredient identification result sequence to obtain a cooking recipe, and the smart terminal transmits the cooking recipe to the human-computer interaction interface.

[0010] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the intelligent image recognition method based on the automatic cooking machine as described above is implemented.

[0011] A storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to execute the intelligent image recognition method based on an automatic cooking machine as described above.

[0012] The beneficial effects of the present invention are: (1) Innovative application of image enhancement technology: Based on conventional image preprocessing, this invention creatively introduces a reconstruction model for image enhancement. Through multi-level and multi-dimensional processing of shallow extraction subunits, deep feature extraction subunits, and feature fusion subunits, this reconstruction model can more accurately capture detailed information in the image, effectively enhance the contrast and clarity of the image, and provide high-quality input data for subsequent recognition. This innovative design significantly improves the accuracy and stability of image recognition, especially in low-light or complex backgrounds, and can still maintain good recognition results.

[0013] (2) Efficient and accurate recognition model: This paper combines the advantages of the U-net neural network and the VGGNet model to construct a comprehensive recognition model that can perform both image segmentation and high-precision recognition. The U-net model, with its excellent image segmentation capabilities, can accurately separate ingredients from the complex kitchen background; while the VGGNet model, with its powerful feature extraction and classification capabilities, can perform high-precision recognition on the segmented images, thereby obtaining an accurate sequence of ingredient recognition results. This combination not only improves the speed and accuracy of recognition, but also enhances the flexibility and adaptability of the system.

[0014] (3) Intelligent cooking recipe matching and interaction: Based on the ingredient recognition result sequence, the present invention can automatically retrieve cooking recipes from the database and transmit the matching cooking recipes to the human-computer interaction interface for users to select. This design not only simplifies the operation process and improves the user experience, but also realizes customized cooking services to meet the taste preferences and dietary needs of different users. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0016] Figure 1 The figure is a flow chart of an intelligent image recognition method based on an automatic cooking machine according to the present invention. DETAILED DESCRIPTION

[0017] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0018] See also Figure 1 ,An intelligent image recognition method based on automatic cooking machine, The food image data is acquired by a built-in visual sensor in the feed bin of the automatic cooking machine, and image preprocessing data is calculated based on the food image data by a preprocessing model; Obtaining a food identification result sequence by calculating the image preprocessing data through a recognition model; The database is traversed according to the ingredient recognition result sequence to obtain a cooking recipe, and the intelligent terminal transmits the cooking recipe to the human-computer interaction interface.

[0019] This embodiment also provides a cucumber and egg recognition method based on intelligent image recognition of an automatic cooking machine, comprising the following steps: Step 1: The staff pours cucumbers and eggs into the feed bin. The built-in gravity sensor in the feed bin obtains real-time gravity data, presets a gravity threshold, and determines whether the real-time gravity data is greater than the gravity threshold. If yes, step 2 is executed; otherwise, no processing is performed. Step 2: The intelligent terminal generates an adaptive signal to control the built-in visual sensor in the feed bin of the automatic cooking machine to obtain food image data, and calculates the image preprocessing data based on the food image data through the preprocessing model; Step 3: Obtain cucumber recognition results and egg recognition results based on the image preprocessing data through the recognition model calculation; Step 4: First, search the database based on the cucumber recognition result to obtain recipes containing cucumber ingredients and store them to obtain sequence 1. Then, search the database based on the egg recognition result to obtain recipes containing egg ingredients and store them to obtain sequence 2. Step 5: Recipes such as cucumber scrambled eggs, cucumber egg soup, and cucumber egg custard are obtained through union-find calculation based on sequence one and sequence two and transmitted to the human-computer interaction interface for manual recipe selection.

[0020] Specifically, the preprocessing model calculation step includes: Normalizing the food image data using a linear function conversion method to obtain normalized data; Gray-scaling the normalized data using a weighted average method to obtain gray-scale data; Performing noise reduction processing on the grayscale data by median filtering to obtain noise-reduced data; The image preprocessing data is obtained by calculating a reconstruction model according to the noise reduction data.

[0021] Specifically, the reconstruction model includes a shallow feature extraction subunit, a deep feature extraction subunit, and a feature fusion subunit. The specific calculation steps include: Obtain shallow features by calculating the shallow feature extraction subunit according to the noise reduction data; A deep feature extraction subunit is used to calculate the denoised data to obtain deep sub-features, and a 1*1 convolution kernel is used to perform feature fusion on the deep sub-features to obtain deep features; The image preprocessing data is obtained by calculation through a feature fusion subunit according to the shallow features and the deep features.

[0022] In this embodiment, the shallow feature extraction subunit is composed of a 3*3 convolution kernel, the deep feature extraction subunit is composed of three multi-branch feature fusion modules, the multi-branch feature fusion module is composed of two cascaded attention modules, and the feature fusion subunit is composed of two 3*3 deconvolution kernels and non-parametric sub-pixel convolution kernels.

[0023] Specifically, the recognition model calculation step includes: Perform image segmentation using a U-net neural network based on the image preprocessing data to obtain an image of the area to be identified; Obtaining a recognition result by calculating the image of the area to be recognized through the VGGNet model; Each of the identification results is stored to obtain the food identification result sequence.

[0024] In this embodiment, the VGGNet model consists of a convolutional layer, an activation function, a pooling layer, a fully connected layer, and a Softmax layer; the U-net neural network consists of an encoder, a decoder, a skip connection, a convolutional layer, a pooling layer, an upsampling layer, a batch normalization, a loss function, and an output layer.

[0025] Specifically, traversing the database according to the ingredient identification result sequence to obtain a cooking recipe specifically includes: S201: Preset a traversal pointer, and sequentially traverse the food identification result sequence according to the traversal pointer to obtain a search identification result; S202: Traversing the database according to the search and identification results to obtain a recipe subsequence; S203: Determine whether the traversal pointer is empty. If yes, execute step S204; if no, repeat steps S201-S203; S204: Obtain the cooking recipe through a union-find calculation based on the recipe subsequence.

[0026] An intelligent image recognition system based on an automatic cooking machine includes: an acquisition unit, a recognition unit, and a query unit; The acquisition unit is used to obtain food image data through a built-in visual sensor in the feed bin of the automatic cooking machine, and obtain image preprocessing data based on the food image data through calculation using a preprocessing model; The recognition unit is configured to calculate a food identification result sequence based on the image preprocessing data through a recognition model; The query unit is used to traverse the database according to the ingredient identification result sequence to obtain a cooking recipe, and the smart terminal transmits the cooking recipe to the human-computer interaction interface.

[0027] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the intelligent image recognition method based on the automatic cooking machine as described above is implemented.

[0028] A storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to execute the intelligent image recognition method based on an automatic cooking machine as described above.

[0029] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0030] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0031] The program code included in the computer-readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF or the like, or any suitable combination thereof. The computer program code for performing the operation of the present invention can be written in one or more programming languages or a combination thereof, and the programming language includes an object-oriented programming language such as Java, Smalltalk, C++, and also includes a conventional procedural programming language such as "C" language or similar programming language. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, utilizing an Internet service provider to connect through the Internet).

[0032] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. An intelligent image recognition method based on an automatic cooking machine, characterized in that: include: The food image data is acquired by a built-in visual sensor in the feed bin of the automatic cooking machine, and image preprocessing data is calculated based on the food image data by a preprocessing model; Obtaining a food identification result sequence by calculating the image preprocessing data through a recognition model; The database is traversed according to the ingredient recognition result sequence to obtain a cooking recipe, and the intelligent terminal transmits the cooking recipe to the human-computer interaction interface.

2. The intelligent image recognition method based on the automatic cooking machine according to claim 1 is characterized in that: The pre-processing model calculation step includes: Normalizing the food image data using a linear function conversion method to obtain normalized data; Gray-scaling the normalized data using a weighted average method to obtain gray-scale data; Performing noise reduction processing on the grayscale data by median filtering to obtain noise-reduced data; The image preprocessing data is obtained by calculating a reconstruction model according to the noise reduction data.

3. The intelligent image recognition method based on the automatic cooking machine according to claim 2 is characterized in that: The reconstruction model includes a shallow feature extraction subunit, a deep feature extraction subunit, and a feature fusion subunit. The specific calculation steps include: Obtain shallow features by calculating the shallow feature extraction subunit according to the noise reduction data; A deep feature extraction subunit is used to calculate the denoised data to obtain deep sub-features, and a 1*1 convolution kernel is used to perform feature fusion on the deep sub-features to obtain deep features; The image preprocessing data is obtained by calculation through a feature fusion subunit according to the shallow features and the deep features.

4. The intelligent image recognition method based on the automatic cooking machine according to claim 1 is characterized in that: The recognition model calculation step includes: Perform image segmentation using a U-net neural network based on the image preprocessing data to obtain an image of the area to be identified; Obtaining a recognition result by calculating the image of the area to be recognized through the VGGNet model; Each of the identification results is stored to obtain the food identification result sequence.

5. The intelligent image recognition method based on the automatic cooking machine according to claim 1 is characterized in that: The step of traversing the database according to the ingredient identification result sequence to obtain a cooking recipe specifically includes: S201: Preset a traversal pointer, and sequentially traverse the food identification result sequence according to the traversal pointer to obtain a search identification result; S202: Traversing the database according to the search and identification results to obtain a recipe subsequence; S203: Determine whether the traversal pointer is empty. If yes, execute step S204; if no, repeat steps S201-S203; S204: Obtain the cooking recipe through a union-find calculation based on the recipe subsequence.

6. An intelligent image recognition system based on an automatic cooking machine, characterized in that: include: Collection unit, identification unit, query unit; The acquisition unit is used to obtain food image data through a built-in visual sensor in the feed bin of the automatic cooking machine, and obtain image preprocessing data based on the food image data through calculation using a preprocessing model; The recognition unit is configured to calculate a food identification result sequence based on the image preprocessing data through a recognition model; The query unit is used to traverse the database according to the ingredient identification result sequence to obtain a cooking recipe, and the smart terminal transmits the cooking recipe to the human-computer interaction interface.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the intelligent image recognition method based on the automatic cooking machine as described in any one of claims 1 to 5 is implemented.

8. A storage medium containing computer-executable instructions, characterized in that: When executed by a computer processor, the computer executable instructions are used to execute the intelligent image recognition method based on an automatic cooking machine as described in any one of claims 1 to 5.

Images