Intelligent dish washing machine based on visual identification

Abstract

The invention relates to the technical field of dish-washing machines, in particular to an intelligent dish-washing machine based on visual identification, comprising: an image acquisition module configured to acquire image information of a target space before the dish-washing machine starts a washing program; wherein the target space is defined as an area which is defined by vertically extending upwards from the bottom of an inner container to the bottom face of a bowl basket on the last but one layer in an inner cavity of the dish washing machine, and the space, except the area, of the inner cavity is a non-target space; and the control unit is connected with the image acquisition module, and is configured to process the acquired image information of the target space, generate a corresponding load signal through a preset image recognition algorithm, and match a preset cleaning mode according to the load signal, so that the cleaning efficiency and the cleaning quality are improved, and meanwhile, the waste of water resources is avoided. The intelligent level of the dish-washing machine is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of dishwashers, in particular to an intelligent dishwasher based on visual recognition. BACKGROUND

[0002] Dishwashers are a major category of modern home appliances. The advent of dishwashers has greatly reduced the burden of kitchen labor. With the continuous development of technology, people try to apply image recognition and algorithm models to dishwashers to improve the intelligence of dishwashers, ultimately achieving the goal of improving cleaning effect and user experience and saving water and electricity.

[0003] For example, patent number 2023117925687 provides a dishwasher control method, device, dishwasher and storage medium. Before running the washing program, in response to the dishwasher door being closed, the placement information of the basket support in the dishwasher, such as linear displacement information, is obtained. Based on the placement information of the basket support, it is determined whether the tableware is placed regularly, and a prompt message is generated when the tableware is not placed regularly to remind the user to reposition the tableware, so as to avoid the occurrence of dead corners and incomplete cleaning due to irregular tableware placement.

[0004] Patent number 2024102350624 provides a dishwasher for processing washing appliances, which proposes to detect washing appliance characteristics through a washing appliance recognition system, such as using visible light to identify different types of washing appliances and the number of washing appliances, or using infrared to identify temperature information, and then activating the appropriate washing appliance processing program to help select the optimal preset washing program of the dishwasher.

[0005] Patent number 2024118168640 provides a running control method and device for a dishwasher. By obtaining a first phase angle image, a second phase angle image, a third phase angle image, a fourth phase angle image, and pressure data of a pressure sensor arranged at the bottom of the dishwasher as initial parameters, and combining an oil stain detection model and a density detection model, the oil stain detection result and the density detection result of the tableware are obtained. These optimized cleaning parameters are used.

[0006] Patent number 2025106167425 provides a dishwasher and its control method, control device and readable storage medium. The recognition module arranged on the inner tank obtains the dirty position of the tableware. The controller can control the first spray arm assembly to operate according to the preset trajectory and the preset spray angle based on the dirty position of the tableware. Real-time pictures of the tableware during the cleaning process are obtained and compared with the pictures recorded at the previous time. When it is determined that the similarity of the dirty color at the same position in the real-time picture and the picture recorded at the previous time is higher than a preset value, the first spray arm assembly is controlled to operate to clean the next dirty position.

[0007] However, these solutions have obvious defects in practical application. First, the image recognition component is complex and often requires multiple cameras, which is almost impossible to achieve in the limited space of a dishwasher. Second, the calculation process is complex, requiring high computing power and signal transmission, which is difficult to apply in the household appliance scenario. Third, after obtaining the information, only the pre-set program can be optimized, and the physical process cannot be effectively regulated, resulting in effective effects.

[0008] Patent No. 2025105012554 provides a washing control method, device, dishwasher equipment, computer equipment and medium. By collecting the image of the dishwasher equipment placement layer and automatically identifying the placement object on the placement layer, the target object is identified based on a preset image processing algorithm. The dirtiness of the target object is determined according to the characteristic parameters including type, quantity and size, the placement layer that needs to be cleaned is automatically divided, and the cleaning component is controlled to clean the area that needs to be cleaned. This technical solution can realize layered cleaning based on image processing, which realizes a regulation of the physical process.

[0009] However, the data processing amount of this scheme is still large, and the occlusion between the layers of the dishwasher is serious, resulting in low accuracy, and only layered cleaning can be achieved, which is not an ideal intelligent solution. SUMMARY

[0010] To solve the problem of complex image recognition scheme and insufficient regulation effect in existing dishwasher technology, the present disclosure provides an intelligent dishwasher based on visual recognition, which can achieve better intelligent regulation at a lower cost.

[0011] The technical solution adopted by the present disclosure to solve the above technical problems is: an intelligent dishwasher based on visual recognition, comprising: An image acquisition module configured to acquire image information of a target space before the dishwasher starts a washing program; Wherein, the target space is defined as the area within the dishwasher cavity vertically extending from the bottom of the inner tank to the bottom surface of the second-to-last layer of the basket, and the space outside the area is the non-target space; A control unit connected to the image acquisition module, configured to process the acquired image information of the target space, and generate a corresponding load signal through a preset image recognition algorithm, and match a preset cleaning mode according to the load signal; A spray arm assembly for cleaning the dishes to be cleaned in the target space; A drive assembly connected to the spray arm assembly for driving the spray arm assembly to rotate; Among them, the control unit is connected to the driving component, and the control unit is configured to output a control instruction to the driving component according to the generated load signal to drive the spray arm component to perform cleaning.

[0012] As a preferred embodiment, the spray arm component includes a main spray arm and an auxiliary spray arm connected to the main spray arm.

[0013] As a preferred embodiment, the driving component is a motor, and the driving shaft of the motor is connected to the main spray arm.

[0014] As a preferred embodiment, a target bowl basket is provided in the target space, and the target bowl basket is divided into a first area, a second area, a third area and a fourth area. The areas of the four areas are equal and distributed in a "field" shape; among them, the first area and the fourth area are arranged diagonally; the second area and the third area are arranged diagonally.

[0015] As a preferred embodiment, the load of each area includes a first threshold, a second threshold and a third threshold, and the cleaning modes include efficient cleaning, key cleaning and normal cleaning; Trigger conditions for efficient cleaning: There is a load in any one of the first area, the second area, the third area and the fourth area; Or there are loads in any two diagonally arranged areas; Trigger conditions for key cleaning: There are loads in any two adjacent areas, and the loads are at different thresholds; Or there are loads in more than three areas, and the load in any one area is at the third threshold; Or there are loads in more than three areas, and the loads in any two diagonally arranged areas are both at the third threshold; Trigger conditions for normal cleaning: There are loads in any two adjacent areas, and the loads are at the same threshold; Or there are loads in more than three areas, and the load thresholds of all areas are the same; Or there are loads in more than three areas, and the loads in any two adjacent areas are both at the third threshold.

[0016] As a preferred embodiment, the first threshold means that the number of dishes to be cleaned is 0, the second threshold means that the number of dishes to be cleaned is 1-3, and the third threshold means that the number of dishes to be cleaned is 4 or more.

[0017] As a preferred implementation manner, the cleaning time of the general cleaning is t, the spray arm assembly cleans the whole target basket in the general cleaning mode; the cleaning time of the high-efficiency cleaning is 50%t-t, the spray arm assembly only cleans the area where the dishes to be cleaned exist in the high-efficiency cleaning mode; the cleaning time of the key cleaning is t, the spray arm assembly cleans the area where the dishes to be cleaned are more for t1 and the area where the dishes to be cleaned are less for t2, where t1>t2 and t1+t2=t.

[0018] As a preferred implementation manner, t1=t2+20%t2.

[0019] As a preferred implementation manner, the image recognition algorithm model is a deep neural network model; the deep neural network model is a convolutional neural network or a generative adversarial network or a transformer or a hybrid expert model.

[0020] As a preferred implementation manner, further comprising a water distribution valve assembly, the water distribution valve assembly comprising a valve seat and a valve sheet, the main spray arm being connected with the valve seat, the valve seat being provided with a water distribution cavity, the valve sheet being arranged at the opening of the water distribution cavity, the valve sheet being connected with a driving member for driving the rotation thereof, the driving member being connected with the control unit.

[0021] As a preferred implementation manner, the dishwasher comprises a shell and a door plate hinged to the shell, and an inner container arranged in the interior of the shell; the image acquisition module is a camera, and the optical view field of the camera covers the whole target space without blind area.

[0022] In some embodiments, the camera is one, and the camera is arranged on the inner surface of the door plate on the side facing the inner container; the door plate width is L, in the horizontal direction, the vertical distance between the camera optical axis center point and the left edge of the door plate is L1, the vertical distance between the camera optical axis center point and the right edge of the door plate is L2, L1≥0.25L; L2≥0.25L; in the vertical direction, the camera is located in the axial space region between the upper surface of the target basket and the lower surface of the adjacent upper basket.

[0023] In some embodiments, the camera is two, and the two cameras are respectively arranged on two opposite side walls in the inner container, the installation inclination angle of each of the two cameras is θ, the axial distance from the camera to the center point of the target basket is H, and the horizontal distance from the camera to the center point of the target basket is D, which satisfy θ=arctan(H / D).

[0024] Compared with the existing products, the image acquisition module of the present application only acquires the image of the target space, and the control unit only analyzes the data of the target space image information. Compared with the traditional image acquisition, the image acquisition module of the present application acquires less images, and the control unit also has less amount of image analysis and processing. In this way, it is not necessary to use high-configuration image acquisition module and control unit. With lower-configuration image acquisition module and control unit, images can be effectively acquired and analyzed and processed, which saves more cost. The present application divides the inner cavity of the dishwasher into target space and non-target space. The control unit matches the most suitable cleaning mode for the target space according to the load condition, thereby improving the cleaning efficiency and cleaning quality, avoiding the waste of water resources, and improving the intelligent level of the dishwasher. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 Figure 1 is a structural schematic diagram of an intelligent dishwasher based on visual recognition according to the present application; Figure 2 Figure 2 is a structural schematic diagram of an intelligent dishwasher based on visual recognition according to the present application; Figure 3 Figure 3 is a structural schematic diagram of an intelligent dishwasher based on visual recognition according to the present application; Figure 4 Figure 4 is a structural schematic diagram of the connection between the driving assembly and the main spray arm; Figure 5 Figure 5 is a processing process of the image recognition algorithm.

[0026] In the figure: 1, target space; 2, door panel; 3, camera; 4, main spray arm; 5, driving assembly; 6, driving shaft; 7, water distribution valve assembly. DETAILED DESCRIPTION

[0027] In order for those skilled in the art to better understand the technical solutions of the present disclosure, the present disclosure will be described in detail, clearly and completely in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present disclosure and do not limit the present disclosure.

[0028] Those skilled in the art should understand that in the disclosure of the present application, the terms "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the above terms cannot be understood as a limitation of the present application.

[0029] As Figures 1-5As shown, the present application provides an intelligent dishwasher based on visual recognition, which comprises an image acquisition module, a control unit, a spray arm assembly and a driving assembly 5. The image acquisition module is configured to acquire image information of a target space 1 before the dishwasher starts the washing program; wherein the target space 1 is defined as the area in the dishwasher cavity vertically extending from the bottom of the inner tank to the bottom surface of the second-to-last layer of the basket, and the space in the cavity outside the area is the non-target space (such as Figure 1 As shown, the dashed area is the target space); the control unit is electrically connected with the image acquisition module and is configured to process the acquired image information of the target space 1 and generate a corresponding load signal through a preset image recognition algorithm (the processing process is shown in Figure 5 As shown), and match a preset cleaning mode according to the load signal; the spray arm assembly is used for cleaning the dishes to be cleaned in the target space 1; the driving assembly 5 is connected to the spray arm assembly for driving the spray arm assembly to rotate; the control unit is electrically connected with the driving assembly 5, and the control unit is configured to output a control instruction to the driving assembly 5 according to the generated load signal to drive the spray arm assembly to clean. In the prior art, the acquisition range of the image recognition component is usually wide (acquiring the image of the entire dishwasher cavity), which makes the number of image recognition components or the structure more complex, and the data processing capacity of the control unit is also higher, which not only requires more space in the cavity, but also has higher requirements for the configuration of the image recognition component and the control unit, which means an increase in cost. In addition, the current directional cleaning mostly sets the dishes to be cleaned in the lower basket as the directional cleaning object, and the acquisition range of the image recognition component is basically the entire cavity of the dishwasher, which has little significance for the image acquisition in the space outside the lower basket, but increases the data processing amount and cost. The image acquisition module of the present application only acquires the image of the target space 1, and the control unit only analyzes the data of the image information of the target space 1. Compared with the traditional image acquisition, the image acquisition module of the present application acquires less image, and the control unit has less image analysis and processing amount. In this way, it is not necessary to use high-configuration image acquisition module and control unit, and low-configuration image acquisition module and control unit can effectively acquire and analyze the image, which saves cost. In addition, since the present application only needs to acquire the image of the target space 1, it is not necessary to set too many image acquisition modules, which greatly relieves the congestion in the cavity of the dishwasher. The present application divides the cavity of the dishwasher into the target space 1 and the non-target space, and the control unit matches the most suitable cleaning mode for the target space 1 according to the load condition, thereby improving the cleaning efficiency and cleaning quality, avoiding the waste of water resources, and improving the intelligent level of the dishwasher.

[0030] The spray arm assembly comprises a main spray arm 4 and an auxiliary spray arm connected to the main spray arm 4. Specifically, the main spray arm 4 comprises a spray arm part and a connecting part, the connecting part is used to connect the water distribution valve assembly 7, and the inner cavity of the spray arm part and the inner cavity of the connecting part are in communication with each other. The auxiliary spray arm comprises at least two auxiliary spray heads arranged on the spray arm part, and the inner cavities of the auxiliary spray heads and the inner cavity of the spray arm part are in communication with each other. The auxiliary spray head is provided with a plurality of spray holes, and the spray arm part can be provided with or not provided with spray holes. When the auxiliary spray head is two, the spray arm part is provided with spray holes, and when the number of auxiliary spray heads is three or more, the spray arm part does not need to be provided with spray holes. More specifically, the main spray arm 4 is a long one-word-shaped spray arm, and the auxiliary spray head is a short one-word-shaped or three-pronged shape, and the specific shape of the auxiliary spray head is not limited here. In order to make the spray arm assembly more stable during rotation, the auxiliary spray head is symmetrically arranged on the spray arm part.

[0031] Specifically, the driving assembly 5 is a motor, and the driving shaft 6 of the motor is connected to the connecting part of the main spray arm 4 through a spline. It should be noted that the main spray arm 4 rotates under the drive of the driving assembly 5, and the auxiliary spray arm rotates with the main spray arm 4, and at the same time, the auxiliary spray arm also rotates under the action of water pressure, that is, when the auxiliary spray arm sprays water, it will rotate relative to the main spray arm 4.

[0032] Specifically, a target bowl basket is arranged in the target space 1. The target bowl basket is divided into a first area, a second area, a third area and a fourth area. The four areas have equal areas and are distributed in a "field" shape. Among them, the first area and the fourth area are arranged diagonally; the second area and the third area are arranged diagonally. It should be noted that the first area, the second area, the third area and the fourth area here can be achieved by physical partition methods, such as: detachable cross, or by the image algorithm in the control unit. The loads of each area include a first threshold, a second threshold and a third threshold. The first threshold means that the number of dishes to be cleaned is 0. The second threshold means that the number of dishes to be cleaned is 1-~3. The third threshold means that the number of dishes to be cleaned is 4 or more. The cleaning modes include efficient cleaning, key cleaning and normal cleaning. The trigger condition for efficient cleaning: There is a load in any one of the first area, the second area, the third area and the fourth area; or there are loads in any two diagonally arranged areas. The trigger condition for key cleaning: There are loads in any two adjacent areas, and the loads are in different thresholds; or there are loads in more than three areas, and the load in any one area is at the third threshold; or there are loads in more than three areas, and the loads in any two diagonally arranged areas are both at the third threshold. The trigger condition for normal cleaning: There are loads in any two adjacent areas, and the loads are in the same threshold; or there are loads in more than three areas, and the load thresholds of all areas are the same; or there are loads in more than three areas, and the loads in any two adjacent areas are both at the third threshold. It can be understood that the control unit processes the images collected by the image acquisition module and generates corresponding load signals, and the control unit will match the preset cleaning mode according to the load signals, that is, the control unit will automatically match the corresponding cleaning mode according to the load conditions of each area. For example: The load signals of the first area are A0, A1, A2, corresponding to the first threshold, the second threshold and the third threshold respectively. When the number of dishes to be cleaned in the first area is 0, the load signal is A0; when the number of dishes to be cleaned in the first area is 1-~3, the load signal is A1; when the number of dishes to be cleaned in the first area is 4 or more (including 4), the load signal is A2. And so on, the load signals of the second area are B0, B1, B2; the load signals of the third area are C0, C1, C2; the load signals of the fourth area are D0, D1, D2.

[0033] Specifically, the cleaning time of the general cleaning is t, in the general cleaning mode, the spray arm assembly cleans the whole target basket, and the cleaning time of each area is 25% t; the cleaning time of the high-efficiency cleaning is 50% t-t, preferably, the cleaning time of the high-efficiency cleaning is 50% t-80% t, in the high-efficiency cleaning mode, the spray arm assembly only cleans the area where the tableware to be cleaned exists; the cleaning time of the key cleaning is t, in the key cleaning mode, the cleaning time of the area where the tableware to be cleaned is more is t1, and the cleaning time of the area where the tableware to be cleaned is less is t2, wherein t1>t2, t1+t2=t. As a preferred, t1=t2+20% t2.

[0034] By coupling control of space and cleaning time, the cleaning efficiency is improved. Specifically, in the high-efficiency cleaning mode, the control unit controls the driving assembly 5 to operate, the driving assembly 5 drives the main spray arm 4 to rotate to the area with load, the main spray arm 4 is kept below the area with load, and the auxiliary spray arm rotates to spray and clean the area with load under the action of water pressure. In the key cleaning mode, the control unit controls the driving assembly 5 to operate, the driving assembly 5 drives the main spray arm 4 to rotate, and in this cleaning mode, the cleaning time of the spray arm assembly for the high-load area is greater than that for the low-load area. In the general cleaning mode, the cleaning time of each area is the same, and each area is cleaned without difference. By dividing the four areas of the target basket and matching different cleaning modes according to the load of each area, the cleaning efficiency and cleaning cleanliness are effectively improved, and the waste of water resources is reduced.

[0035] For example, if the first area and the fourth area have load (the load threshold value can be the same or different), the spray arm assembly only cleans the first area and the fourth area, and the total cleaning time is 50% t-80% t, which is high-efficiency cleaning. If the first area, the second area and the fourth area have load, for example, the number of tableware to be cleaned in the first area is 2, the number of tableware to be cleaned in the second area is 1, and the number of tableware to be cleaned in the fourth area is 5, the cleaning time of the spray arm assembly for the first area and the fourth area is t1, and the cleaning time for the second area is t2, which is key cleaning. It should be noted that when the spray arm assembly sprays the fourth area, the first area, which is the diagonal area of the fourth area, will also be sprayed. Therefore, although the load of the first area is small, the cleaning time of the first area is the same as that of the fourth area which has more load. If the first area, the second area and the third area all have load, for example, the number of tableware to be cleaned in the first area is 2, the number of tableware to be cleaned in the second area is 1, and the number of tableware to be cleaned in the third area is 3, the spray arm assembly cleans each area, and the cleaning time is the same.

[0036] Preferably, the image recognition algorithm model is a deep neural network model. Specifically, the deep neural network algorithm model is a convolutional neural network (CNN), a generative adversarial network (GAN), a transformer, or a hybrid expert model (MoE).

[0037] In one embodiment of this disclosure, a water distribution valve assembly 7 is further included. The water distribution valve assembly 7 includes a valve seat and a valve plate. The main spray arm 4 is connected to the valve seat. The valve seat is provided with a water distribution chamber. The valve plate is located at the opening of the water distribution chamber. The valve plate is connected to a driving component for driving its rotation. The driving component is electrically connected to the control unit. Specifically, the driving component is a motor. The output shaft of the motor is connected to the valve plate to drive the valve plate to rotate. It is understood that the dishwasher is also provided with an inner water pipe, a middle spray arm, and an upper spray arm. The middle spray arm and the upper spray arm are both connected to the inner water pipe. The middle spray arm and the upper spray arm are used to clean the dish rack in non-target areas. The inner water pipe and the main spray arm 4 are both connected to the water distribution chamber. The driving component controls whether water flows through the inner water pipe and / or the main spray arm 4 by controlling the rotation of the valve plate. The connection methods of the upper spray arm, the middle spray arm, and the inner water pipe, as well as the connection between the inner water pipe, the main spray arm 4, and the valve seat, are all prior art and will not be described in detail here.

[0038] The dishwasher includes a housing and a door panel 2 hinged to the housing, with the inner liner located inside the housing; the image acquisition module is a camera 3, and the optical field of view of the camera 3 covers the entire target space 1 without blind spots.

[0039] like Figure 3 As shown, in some embodiments, there is only one camera 3, which is mounted on the inner surface of the door panel 2 facing the inner liner. Let the width of the door panel 2 be L. In the horizontal direction, the vertical distance between the center point of the optical axis of the camera 3 and the left edge of the door panel 2 is L1, and the vertical distance between the center point of the optical axis of the camera 3 and the right edge of the door panel 2 is L2, where L1 ≥ 0.25L and L2 ≥ 0.25L. In the vertical direction, the center point of the optical axis of the camera 3 is located within the axial space between the upper surface of the target bowl basket and the lower surface of the adjacent upper bowl basket. This embodiment uses only one camera 3, reducing the number of components, lowering hardware costs and failure rates. The camera 3 is mounted on the door panel 2 without occupying space within the inner liner and is easy to maintain. Placing the camera 3 at the aforementioned position on the door panel 2 can eliminate recognition errors caused by perspective distortion and improve image recognition accuracy. The camera 3 is confined within the axial space between the upper surface of the target bowl basket and the lower surface of the adjacent upper bowl basket, ensuring that it is not obstructed by upper tableware or blocked by lower tableware due to its low position, thus guaranteeing the integrity of image acquisition.

[0040] like Figure 2As shown, in some embodiments, the camera 3 is two, two cameras 3 are respectively arranged on two opposite side walls in the inner container, the installation angle of each of the two cameras 3 is θ, the axial distance from the camera 3 to the center point of the target basket is H, and the horizontal distance from the camera 3 to the center point of the target basket is D, and the three satisfy θ=arctan(H / D). Specifically, θ is the angle between the optical axis of the camera 3 and the horizontal plane, H is the vertical distance from the optical center of the camera 3 to the center point of the target basket, and D is the horizontal distance from the projection point of the optical center of the camera 3 on the horizontal plane to the center point of the target basket. Using two cameras 3 can significantly improve the accuracy of image recognition and the accuracy of depth information; the positional relationship θ=arctan(H / D) can ensure that the field of view of the two cameras 3 realizes the maximum range of coverage in the target space 1, while avoiding image distortion and recognition blind area caused by too large or too small angle.

[0041] The above has carried out the detailed introduction to the present application, the principle and implementation mode of the present application are described in this paper by applying specific examples, the above embodiment is only used to help understanding the present application and core idea. It should be pointed out that for ordinary skilled in the art, without departing from the principle of the present application, the present application can be improved and modified, these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A vision recognition-based intelligent dishwasher, characterized by, Comprising: An image acquisition module, configured to acquire image information of a target space (1) before the dishwasher starts a washing program; Wherein, the target space (1) is defined as the area in the dishwasher cavity vertically extending upward from the bottom of the inner tank to the bottom surface of the second-to-last layer of the bowl basket, and the inner cavity space outside this area is a non-target space; A control unit, connected to the image acquisition module, configured to process the acquired image information of the target space (1), generate a corresponding load signal through a preset image recognition algorithm, and match a preset cleaning mode according to the load signal; A spray arm assembly, used to clean the tableware to be cleaned in the target space (1); A drive assembly (5), connected to the spray arm assembly, used to drive the spray arm assembly to rotate; Wherein, the control unit is connected to the drive assembly (5), and the control unit is configured to output a control instruction to the drive assembly (5) according to the generated load signal to drive the spray arm assembly to perform cleaning. 2.The vision recognition-based intelligent dishwasher according to claim 1, characterized in that, The spray arm assembly includes a main spray arm (4) and an auxiliary spray arm connected to the main spray arm (4). 3.The vision recognition-based intelligent dishwasher according to claim 2, characterized in that, The drive assembly (5) is a motor, and the drive shaft (6) of the motor is connected to the main spray arm (4). 4.The vision recognition based intelligent dishwasher according to claim 1, characterized in that, A target bowl basket is provided in the target space (1), and the target bowl basket is divided into a first area, a second area, a third area, and a fourth area. The four areas have equal areas and are distributed in a "field" shape; wherein, the first area and the fourth area are arranged diagonally; the second area and the third area are arranged diagonally. 5.The vision recognition based intelligent dishwasher according to claim 4, characterized in that, The load of each area includes a first threshold, a second threshold, and a third threshold, and the cleaning modes include efficient cleaning, key cleaning, and normal cleaning; Trigger conditions for efficient cleaning: There is a load in any one of the first area, the second area, the third area, and the fourth area; Or there are loads in any two diagonally arranged areas; Trigger conditions for key cleaning: There are loads in any two adjacent areas, and the loads are in different thresholds; Or there are loads in more than three areas, and the load in any one area is at the third threshold; Or there are loads in more than three areas, and the loads in any two diagonally arranged areas are both at the third threshold; Trigger conditions for normal cleaning: There are loads in any two adjacent areas, and the loads are in the same threshold; Or there are loads in more than three areas, and the load thresholds of all areas are the same; Or there are loads in more than three areas, and the loads in any two adjacent areas are both at the third threshold. 6.The vision recognition based intelligent dishwasher according to claim 5, characterized in that, The first threshold means that the number of tableware to be cleaned is 0, the second threshold means that the number of tableware to be cleaned is 1 - 3, and the third threshold means that the number of tableware to be cleaned is 4 or more. 7.The vision recognition based intelligent dishwasher according to claim 5, characterized in that, The cleaning time of the normal cleaning is t. In the normal cleaning mode, the spray arm assembly cleans the entire target bowl basket; the cleaning time of the efficient cleaning is 50%t - t. In the efficient cleaning mode, the spray arm assembly only cleans the area where there is tableware to be cleaned; the cleaning time of the key cleaning is t. In the key cleaning mode, the cleaning time of the area with more tableware to be cleaned is t1, and the cleaning time of the area with less tableware to be cleaned is t2, where t1 > t2 and t1 + t2 = t. 8.The vision recognition based intelligent dishwasher according to claim 7, characterized in that, t1=t2+20%t2. 9.The vision recognition based intelligent dishwasher according to claim 1, wherein, The image recognition algorithm model is a deep neural network model; the deep neural network model is a convolutional neural network or a generative adversarial network or a transformer or a hybrid expert model. 10.The vision recognition based intelligent dishwasher according to claim 2, wherein, The water distribution valve assembly (7) comprises a valve seat and a valve piece, the main spray arm (4) is connected with the valve seat, the valve seat is provided with a water distribution cavity, the valve piece is arranged at the opening of the water distribution cavity, the valve piece is connected with a driving member for driving the rotation thereof, and the driving member is connected with the control unit. 11.The vision recognition based intelligent dishwasher according to claim 1, wherein, The dishwasher comprises a shell and a door panel (2) hinged to the shell, an inner container is arranged in the interior of the shell; the image acquisition module is a camera (3), and an optical view field of the camera (3) covers the whole target space (1) without blind area. 12.The vision recognition based intelligent dishwasher according to claim 11, characterized in that, The camera (3) is one, and the camera (3) is arranged on an inner surface of the door panel (2) on a side facing the inner container; the door panel (2) has a width of L, in the horizontal direction, a vertical distance between a center point of an optical axis of the camera (3) and a left edge of the door panel (2) is L1, a vertical distance between the center point of the optical axis of the camera (3) and a right edge of the door panel (2) is L2, L1 is greater than or equal to 0.25L, L2 is greater than or equal to 0.25L; in the vertical direction, the camera (3) is located in an axial space region between an upper surface of a target basket and a lower surface of an adjacent upper basket. 13.The vision recognition based intelligent dishwasher according to claim 11, wherein, The camera (3) is two, and the two cameras (3) are respectively arranged on two opposite side walls of the inner container, an installation inclination angle of each of the two cameras (3) is θ, an axial distance from the camera (3) to a center point of a target basket is H, and a horizontal distance from the camera (3) to the center point of the target basket is D, and the three satisfy θ=arctan(H / D).

Images