A shared intelligent electric oven controlled by graphic recognition technology

Abstract

The present invention relates to the technical field of electric ovens, a shared intelligent electric oven controlled based on graphic recognition technology, comprising an electric oven table main body, a controller with multiple processors and storages built in, and a communicator connected to a cloud service terminal through signals, the controller is connected to the communicator signal, a target detection and recognition module and a target recognition result processing module are arranged in the controller, a support frame is fixedly connected to the top of the electric oven table, a first high-definition camera is fixedly connected to the inner top wall of the support frame, a tilting block is fixedly connected to the side wall of the support frame, a second high-definition camera is fixedly connected to the outside of the tilting block, and the first high-definition camera captures an object picture. The shared intelligent electric oven controlled based on graphic recognition technology automatically controls the barbecue equipment by recognizing the image of the food, avoiding the equipment safety caused by the outdoor users roasting improper food.

Description

Technical Field

[0001] The present invention relates to the technical field of electric ovens, and in particular to a shared intelligent electric oven controlled based on graphic recognition technology. Background Art

[0002] Current smart electric baking products have a variety of functions. Through the application of intelligent technologies, such as temperature control, time setting, automatic preheating, etc., they provide users with an intelligent baking experience, making the baking process easier and simpler, and can meet the needs of different groups of people, such as students, office workers, families, etc. In addition, various sharing technologies can also meet the needs of outdoor sharing, so that people do not need to buy expensive electric oven equipment, they can enjoy the fun of outdoor grilling.

[0003] However, the outdoor maintenance and management of shared public intelligent electric oven equipment manufactured according to the existing patented technology requires a lot of manpower and material resources to ensure the normal operation and safe use of the equipment. At the same time, the user's usage habits and behaviors also need to be standardized and restricted. Some users will use tableware and food that are prohibited from being placed in the electric oven, resulting in damage to the equipment or waste of resources. Therefore, there is an urgent need for a shared intelligent electric oven controlled based on graphic recognition technology to solve the above problems. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the shortcomings of the prior art, the present invention provides a shared intelligent electric oven controlled based on graphic recognition technology, which has the advantages of being able to identify the safety of the items placed in it, and solves the problem that some users use tableware and food that are prohibited from being placed in the electric oven, causing damage to the equipment or wasting resources.

[0006] (II) Technical solution

[0007] The technical solution of the present invention to solve the above technical problems is as follows: a shared intelligent electric oven controlled based on graphic recognition technology, comprising an electric grill main body, a controller with multiple processors and storages built in, and a communicator connected to a cloud service terminal through signals, characterized in that: the controller is connected to the communicator signal, a target detection and recognition module and a target recognition result processing module are arranged in the controller, an electric control box is fixedly connected to the bottom of the electric grill main body, a loudspeaker connected to the signal of the target recognition processing module is fixedly connected to the inside of the electric control box, a support frame is fixedly connected to the top of the electric grill, and a signal signal is fixedly connected to the inner top wall of the support frame. There is a first high-definition camera, a tilting block is fixedly connected to the side wall of the support frame, and a second high-definition camera is fixedly connected to the outside of the tilting block. The first high-definition camera captures object pictures, and the first high-definition camera pre-processes the collected object picture data and sends it to the cloud service terminal through the communicator. The cloud service terminal uses the CSPDarknet backbone network to extract image features, locates and classifies the target through the target detection model, removes duplicate recognition results by performing non-maximum suppression, and then obtains the optimal recognition result by post-processing the confidence, and transmits the data result to the target recognition result processing module.

[0008] The beneficial effects of the present invention are:

[0009] This shared smart electric oven, which is controlled based on graphic recognition technology, automatically controls the barbecue equipment through image recognition of food ingredients, allowing it to work safely. It can provide people with safer and more convenient food baking equipment in public places, avoiding equipment safety risks caused by outdoor users baking improper food ingredients. In addition, equipment sharing can effectively reduce resource waste.

[0010] Based on the above technical solution, the present invention can also be improved as follows.

[0011] Furthermore, an electric oven is arranged inside the main body of the electric grill, an oil leakage hole is arranged inside the electric oven, and a collecting box is slidably connected to the bottom of the main body of the electric grill.

[0012] Furthermore, a QR code label is fixedly connected to the side wall of the support frame, a display is arranged on the side wall of the electric control box, and a control button connected to the controller signal is arranged on the side wall of the electric control box.

[0013] Furthermore, the controller is fixedly connected to the inside of the electric control box, the communicator is fixedly connected to the inside of the electric control box, and a battery is arranged inside the electric control box.

[0014] Furthermore, the second high-definition camera performs a secondary inspection on the baked items to prevent bad users from placing dangerous objects on the electric oven by deceptive means.

[0015] Furthermore, after receiving the data result, the target recognition result processing module executes the following steps: when it is a heatable item, the announcer is used to remind the user that the user can continue normal operation to complete the heating of the food; when it is a non-heatable item, the announcer is used to remind the user that the item is not suitable for heating and the heating operation is terminated.

[0016] Furthermore, the training method of the target detection model is as follows: by dividing the samples into two types, one is a positive sample, which can be placed in an electric oven item picture and manually labeled information; the other is a negative sample that cannot be placed in an electric oven item picture and manually labeled information, by integrating the manually labeled training data into a data set, by dividing the data set into a training data set and a test training set, using the Yo l ov8 detection algorithm to receive the training data set to obtain a model structure, by comparing with the test training set for evaluation, and optimizing the target detection model.

[0017] Furthermore, the user scans the QR code label through the mobile terminal and synchronizes it to the cloud server for model matching. The user authorizes the use and pays the fee. During the baking process, it is heated after being recognized by the target recognition result processing module. If the user does not take action after the reminder when it cannot be heated, a refund will be initiated and the power supply will be interrupted.

[0018] Furthermore, the object image preprocessing method includes: performing various angle rotation processing, various brightness processing, various chromaticity processing, various contrast processing and various sharpness processing on each object image.

[0019] Furthermore, the method for preprocessing the object images also includes: segmenting the object images, separating the food and cooking utensils, and performing image recognition and classification for each. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the structure of the present invention;

[0021] Figure 2 It is a bottom view of the structure of the present invention;

[0022] Figure 3 It is a cross-sectional view of the structure of the present invention;

[0023] Figure 4 Share payment flow chart for the present invention;

[0024] Figure 5 This is the image recognition flow chart of the present invention.

[0025] In the figure: 1. Electric grill main body; 2. Electric control box; 3. Controller; 4. Communicator; 5. Broadcaster; 6. Battery; 7. Support frame; 8. First high-definition camera; 9. Second high-definition camera; 10. Electric grill; 11. Oil leakage hole; 12. Collection box; 13. QR code label; 14. Display; 15. Control button. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] In the embodiment, Figure 1-5 A shared intelligent electric oven controlled based on graphic recognition technology is provided, the present invention comprises an electric oven main body 1, a controller 3 with multiple processors and storages built in, and a communicator 4 connected to a cloud service terminal through signals, the controller 3 is connected to the communicator 4 by signal, a target detection and recognition module and a target recognition result processing module are arranged in the controller 3, an electric control box 2 is fixedly connected to the bottom of the electric oven main body 1, a loudspeaker 5 connected to the target recognition processing module signal is fixedly connected to the inside of the electric control box 2, a support frame 7 is fixedly connected to the top of the electric oven 1, a first high-definition camera 8 is fixedly connected to the inner top wall of the support frame 7, a tilting block is fixedly connected to the side wall of the support frame 7, and a second high-definition camera 9 is fixedly connected to the outside of the tilting block.

[0028] Specifically, refer to Figure 1 An electric oven 10 is arranged inside the electric grill table body 1, an oil leakage hole 11 is arranged inside the electric grill table body 1, a collecting box 12 is slidably connected to the bottom of the electric grill table body 1, a QR code label 13 is fixedly connected to the side wall of the support frame 7, a display 14 is arranged on the side wall of the electric control box 2, and a control button 15 connected to the signal of the controller 3 is arranged on the side wall of the electric control box 2.

[0029] Specifically, refer to Figure 3 The controller 3 is fixedly connected to the inside of the electric control box 2 , the communicator 4 is fixedly connected to the inside of the electric control box 2 , and a battery 6 is arranged inside the electric control box 2 .

[0030] In this embodiment, in actual use, the first high-definition camera 8 captures the image of the object. The first high-definition camera 8 pre-processes the collected object image data and sends it to the cloud service terminal through the communicator 4. The cloud service terminal uses the CSPDarknet backbone network to extract image features, locates and classifies the target through the target detection model, removes duplicate recognition results by performing non-maximum suppression, and then obtains the optimal recognition result by post-processing the confidence, and transmits the data result to the target recognition result processing module.

[0031] After receiving the data result, the target recognition result processing module executes the following steps: when it is a heatable item, the loudspeaker 5 is used to remind the user that the user can continue normal operation and complete the heating of the food; when it is a non-heatable item, the loudspeaker 5 is used to remind the user that the item is not suitable for heating and the heating operation is terminated.

[0032] It should be noted that the user scans the QR code label 13 through the mobile terminal and synchronizes it to the cloud server for model matching. The user authorizes the use and pays the fee. During the baking process, it is heated after being recognized by the target recognition result processing module. If the user does not process it after the reminder, a refund will be initiated and the power supply will be interrupted.

[0033] The training method of the target detection model is as follows: the samples are divided into two types: positive samples, which can be placed in electric oven pictures and manually labeled information; and negative samples, which cannot be placed in electric oven pictures and manually labeled information. The data trained by manual labeling are integrated into a data set, and the data set is divided into a training data set and a test training set. The Yo l ov8 detection algorithm is used to receive the training data set to obtain the model structure, which is evaluated by comparing it with the test training set to optimize the target detection model.

[0034] The preprocessing methods for the object images include: rotating each object image at multiple angles, performing multiple brightness processing, performing multiple color processing, performing multiple contrast processing, and performing multiple sharpness processing, segmenting the object images, separating food from cooking utensils, and performing image recognition and classification respectively.

[0035] Image recognition-based security control is added to the objects entering the electric grill table body 1. The purpose is to identify objects that are not suitable for entering the electric grill and classify their security levels. For example, objects that are completely prohibited from entering are classified as dangerous, such as plastic tableware, glassware, eggs in shells, grapes, etc., and their entry can be restricted. After entering, an alarm will be issued or the system will not work; for objects that can enter but need to be restricted in time, they will be classified as key attention categories, and the range of the heating time that can be set after they enter will be limited; safe objects are also classified as safe categories, and a wider heating time is given.

[0036] Conventional smart electric ovens are equipped with temperature sensors, which collect temperature information of ingredients during cooking. Together with the second high-definition camera 9, they can perform secondary identification and detection of ingredients, thereby preventing bad users from placing dangerous objects in shared smart ovens by deceptive means, and reducing safety risks caused by humans.

[0037] The controller 3 is connected to the electronic terminal signal through the communicator 4. The electronic terminal can be a mobile phone or a tablet, so that the operator can control the food baking status in the electric grill main body 1 on the electronic terminal.

[0038] Working principle:

[0039] The user places the cooking utensils and food under the first high-definition camera 8. The user scans the QR code label 13 through the electronic terminal, enters the mini program, and synchronizes to the cloud server for model matching. The user authorizes the use and pays the fee. After receiving the command, the controller 3 uses the first high-definition camera 8 to capture the image of the object, and sends it to the cloud service terminal through the communicator 4 after pre-processing. The cloud service terminal uses the CSPDarknet backbone network to extract image features, locates and classifies the target through the target detection model, removes duplicate recognition results by non-maximum suppression, and then obtains the optimal recognition result through confidence post-processing. The data result is transmitted to the target recognition result processing module. After receiving the data result, the target recognition result processing module executes the following steps: when it cannot be heated, a refund is initiated, and the user is warned through the loudspeaker 5 that there are non-heatable items, reminding them not to occupy space for a long time. When it can be heated, heating is started to complete the meal.

[0040] The user can adjust the baking condition through the control button 15, or remotely adjust the baking condition through the electronic terminal. At the same time, when the temperature in the electric oven is abnormal, the second high-definition camera 9 performs secondary recognition and detection of the food being baked according to the image recognition technology, and urgently interrupts the heating, and reminds the user and management personnel on the electronic terminal, and reminds the attention of people around the electric baking table body 1 through the loudspeaker 5.

[0041] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0042] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A shared intelligent electric oven controlled based on graphic recognition technology, comprising an electric oven table body (1), a controller (3) with multiple processors and storage devices built in, and a communicator (4) connected to a cloud service terminal via a signal, characterized in that: The controller (3) is connected to the communicator (4) by signal, and a target detection and recognition module and a target recognition result processing module are arranged in the controller (3); the bottom of the electric grilling table body (1) is fixedly connected to an electric control box (2), the inside of the electric control box (2) is fixedly connected to a speaker (5) connected to the signal of the target recognition and processing module, the top of the electric grilling table (1) is fixedly connected to a support frame (7), the inner top wall of the support frame (7) is fixedly connected to a first high-definition camera (8), the side wall of the support frame (7) is fixedly connected to a tilting block, and the outside of the tilting block is fixedly connected to a second high-definition camera (9); The first high-definition camera (8) captures an image of an object. The first high-definition camera (8) pre-processes the collected image data of the object and then sends it to the cloud service terminal through the communicator (4). The cloud service terminal uses the CSPDarknet backbone network to extract image features, locates and classifies the target through the target detection model, removes duplicate recognition results by performing non-maximum suppression, and then obtains the optimal recognition result by post-processing the confidence, and transmits the data result to the target recognition result processing module.

2. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 1, characterized in that: An electric oven (10) is arranged inside the electric grill table body (1), an oil leakage hole (11) is arranged inside the electric grill table body (10), and a collecting box (12) is slidably connected to the bottom of the electric grill table body (1).

3. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 1, characterized in that: A two-dimensional code label (13) is fixedly connected to the side wall of the support frame (7), a display (14) is arranged on the side wall of the electric control box (2), and a control button (15) connected to the controller (3) signal is arranged on the side wall of the electric control box (2).

4. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 1, characterized in that: The controller (3) is fixedly connected to the inside of the electric control box (2), the communicator (4) is fixedly connected to the inside of the electric control box (2), and a storage battery (6) is arranged inside the electric control box (2).

5. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 2, characterized in that: The second high-definition camera (9) performs a secondary inspection on the baked items to prevent bad users from placing dangerous objects on the electric oven (10) by deceptive means.

6. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 1, characterized in that: After receiving the data result, the target recognition result processing module executes the following steps: when the item is heatable, the loudspeaker (5) is used to remind the user that the user can continue normal operation to complete the heating of the food; when the item is not heatable, the loudspeaker (5) is used to remind the user that the item is not suitable for heating and the heating operation is terminated.

7. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 1, characterized in that: The training method of the target detection model is as follows: by dividing the samples into two types, one of which is a positive sample, which can contain electric oven object pictures and manual annotation information; The other is that negative samples cannot be put into electric oven item pictures and manually annotated information. The manually labeled training data are integrated into a data set, and the data set is divided into a training data set and a test training set. The Yolov8 detection algorithm is used to receive the training data set to obtain the model structure, and the target detection model is optimized by comparing it with the test training set for evaluation.

8. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 1, characterized in that: The user scans the QR code label (13) through a mobile terminal, which is synchronized to the cloud server for model matching. The user authorizes the use and pays the fee. During the baking process, the object recognition result processing module recognizes the object and then heats it. If the object cannot be heated and the user fails to process the object after being reminded, a refund is initiated and the power supply is cut off.

9. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 1, characterized in that: The object image preprocessing method includes: performing various angle rotation processing, various brightness processing, various chromaticity processing, various contrast processing and various sharpness processing on each object image.

10. The shared intelligent electric oven controlled based on graphic recognition technology according to claim 9, characterized in that: The method of preprocessing the object images also includes: segmenting the object images, separating the food from the cooking utensils, and performing image recognition and classification respectively.

Images