Intelligent refrigerator and article identification method

Abstract

The application discloses an intelligent refrigerator and an object identification method. When the gray scale of a to-be-processed image including hand information needs to be adjusted, the gray scale of the to-be-processed image can be adjusted according to the brightness of the environment where the hand is located, and then object identification processing is performed on the adjusted image. In this way, the quality of the image collected under poor lighting conditions is poor, and the object cannot be accurately identified. The object can be accurately identified under different lighting conditions, the influence of the lighting conditions on the object identification effect is eliminated, the accuracy of the object identification is improved, the accuracy of the object management is improved, and the experience of the user is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart home, in particular to a smart refrigerator and an article identification method. BACKGROUND

[0002] For the smart refrigerator, accurate management of food material information is a core function of the smart refrigerator, which can not only provide effective food material information for the user, but also provide effective reference and guidance for the user to store food materials, so as to avoid mold of food materials due to long storage time, thereby improving the experience of the user.

[0003] Therefore, how to accurately identify food materials is a technical problem to be solved by those skilled in the art. SUMMARY

[0004] The embodiments of the present application provide a smart refrigerator and an article identification method, which are used to accurately identify articles.

[0005] In a first aspect, the embodiments of the present application provide a smart refrigerator, comprising:

[0006] an image collector;

[0007] a processor configured to:

[0008] determine an image collected by the image collector and including hand information as a to-be-processed image;

[0009] when it is determined that the to-be-processed image needs to adjust the gray scale, adjust the gray scale of the to-be-processed image according to the brightness of the environment where the hand is located;

[0010] perform article identification processing on the to-be-processed image after adjusting the gray scale.

[0011] In a second aspect, the embodiments of the present application provide an article identification method, comprising:

[0012] determine an image collected and including hand information as a to-be-processed image;

[0013] when it is determined that the to-be-processed image needs to adjust the gray scale, adjust the gray scale of the to-be-processed image according to the brightness of the environment where the hand is located;

[0014] perform article identification processing on the to-be-processed image after adjusting the gray scale.

[0015] The present application has the following advantages:

[0016] The smart refrigerator and the article identification method provided in the embodiment of the present application can adjust the gray scale of the to-be-processed image according to the brightness of the environment where the hand is located when the gray scale of the to-be-processed image including hand information needs to be adjusted, and then perform article identification processing on the adjusted image, so that the quality of the image collected under poor lighting conditions is avoided to be poor and the article cannot be accurately identified, the article can be accurately identified under different lighting conditions, the influence of the lighting condition on the article identification effect is eliminated, the accuracy of the article identification and the accuracy of the article management are improved, and the experience of the user is improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 FIG. 1 is a structural schematic diagram of a smart refrigerator provided in an embodiment of the present application;

[0018] Figure 2 FIG. 2 is a structural schematic diagram of another smart refrigerator provided in an embodiment of the present application;

[0019] Figure 3 FIG. 3 is a schematic diagram of the brightness of the environment where the hand is located provided in an embodiment of the present application;

[0020] Figure 4 FIG. 4 is a schematic diagram of the gray scale of the to-be-processed image provided in an embodiment of the present application;

[0021] Figure 5 FIG. 5 is a flowchart of an article identification method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] The specific implementation of the smart refrigerator and the article identification method provided in the embodiment of the present application will be described in detail below with reference to the drawings. It should be noted that the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0023] The embodiment of the present application provides a smart refrigerator, as shown in FIG. 1, which can include: Figure 1

[0024] an image collector 101;

[0025] a processor 102 configured to:

[0026] determine the image collected by the image collector 101 and including hand information as a to-be-processed image;

[0027] when it is determined that the to-be-processed image needs to adjust the gray scale, adjust the gray scale of the to-be-processed image according to the brightness of the environment where the hand is located; ​

[0028] The image to be processed is subjected to an item recognition process.

[0029] In this way, when the gray scale of the image to be processed including hand information needs to be adjusted, the gray scale of the image to be processed can be adjusted according to the brightness of the environment where the hand is located, and the adjusted image is subjected to an item recognition process, so that the image collected under poor lighting conditions can be avoided to have poor quality and cannot accurately recognize the item, so that the item can be accurately recognized under different lighting conditions, the influence of the lighting condition on the item recognition effect is eliminated, the accuracy of the item recognition is improved, and the accuracy of the item management is improved, thereby improving the experience of the user.

[0030] Optionally, in the embodiment of the present application, the item can include but is not limited to food, and can also include cosmetics, medicines, etc.

[0031] Optionally, in the embodiment of the present application, the processor is specifically configured to:

[0032] determine a gray scale image corresponding to the image to be processed;

[0033] adjust the gray scale of each pixel in the gray scale image according to the brightness of the environment where the hand is located;

[0034] perform gamma correction processing on the adjusted gray scale.

[0035] In the embodiment of the present application, the image to be processed collected by the image collector can be but is not limited to an RGB image, and the corresponding image collector can be an RGB camera, so the processor needs to first convert the RGB image into a gray scale image, so as to facilitate subsequent adjustment of the gray scale.

[0036] In this way, the gray scale of the image to be processed can be adjusted according to the brightness of the environment where the hand is located, so as to avoid affecting the subsequent item recognition result due to the too large or too small brightness of the image to be processed, and further avoid eliminating the influence of the brightness on the item recognition result, thereby facilitating improvement of the accuracy of the item recognition.

[0037] Optionally, in the embodiment of the present application, the processor is specifically configured to:

[0038] perform normalization processing on the adjusted gray scale to obtain a real value corresponding to each gray scale;

[0039] perform gamma correction processing on each real value according to a preset gamma value and a gamma correction formula;

[0040] convert each real value after the gamma correction processing into a gray scale.

[0041] The normalization process and the process of converting the real values after gamma correction into gray scales can be implemented by any method known to those skilled in the art, and are not limited herein.

[0042] It is noted that when the gray scale range is assumed to be 0-255, the range of the real values after normalization is 0-1, and the gray scale range after converting the real values after gamma correction into gray scales is still 0-255.

[0043] In this way, the adjusted gray scales can be gamma corrected to compensate for the adjusted gray scales, thereby improving the accuracy of gray scale adjustment and facilitating the improvement of the result of subsequent article identification processing.

[0044] Specifically, the processor is specifically configured to:

[0045] The real values are gamma corrected by the following formula:

[0046] k(x, y) = h(x, y) 1 / gamma ;

[0047] wherein k(x, y) represents the value after gamma correction of the pixel located at the xth row and yth column, h(x, y) represents the real value corresponding to the gray scale of the pixel located at the xth row and yth column, and gamma represents the gamma value.

[0048] When setting gamma, it can be set as:

[0049] The gamma (hereinafter referred to as g1) used when it is determined that the brightness of the environment where the hand is located is less than the preset lower limit of brightness is different from the gamma (hereinafter referred to as g2) used when it is determined that the brightness of the environment where the hand is located is greater than the preset upper limit of brightness, for example, but not limited to, g1 is set to 2.2, and g2 is set to 0.45.

[0050] Of course, g1 and g2 can also be set to be the same, and can be set according to actual needs and CRT correction coefficients, which are not limited herein.

[0051] In this way, gamma correction can be implemented based on the above correction formula, thereby implementing gamma correction of the adjusted gray scales to improve the accuracy of gray scale adjustment.

[0052] Optionally, in the embodiment of the present application, the processor is specifically configured to:

[0053] when it is determined that the brightness of the environment where the hand is located is less than a preset lower limit value of brightness, the adjusted gray scale of each pixel in the gray scale image is the sum of the gray scale of the pixel in the gray scale image and a preset first constant, and when the adjusted gray scale is greater than a preset upper limit value of gray scale, the upper limit value of gray scale is taken as the adjusted gray scale;

[0054] when it is determined that the brightness of the environment where the hand is located is greater than a preset upper limit value of brightness, the adjusted gray scale of each pixel in the gray scale image is the difference between the gray scale of the pixel in the gray scale image and a preset second constant, and when the adjusted gray scale is less than a preset lower limit value of gray scale, the lower limit value of gray scale is taken as the adjusted gray scale.

[0055] The first constant and the second constant can be set to be the same, of course, can also be set to be different, and the first constant and the second constant can also be set according to actual needs, which are not limited here.

[0056] Similarly, the lower limit value of brightness, the upper limit value of brightness, the upper limit value of gray scale, and the lower limit value of gray scale can be set according to actual needs, for example but not limited to: the lower limit value of brightness is 150 Lux, the upper limit value of brightness is 1000 Lux, the upper limit value of gray scale is 255, and the lower limit value of gray scale is 0, which are not limited here.

[0057] For example, assuming that the gray scale range of the pixel is 0-255, the upper limit value of gray scale is 255, the lower limit value of gray scale is 0, and the first constant and the second constant are both assumed to be 50, when f(x, y) represents the gray scale of the pixel in the gray scale image and g(x, y) represents the adjusted gray scale, then:

[0058] when it is determined that the brightness of the environment where the hand is located is less than a preset lower limit value of brightness, g(x, y) = f(x, y) + 50;

[0059] wherein if g(x, y) is greater than 255, g(x, y) = 255.

[0060] Or, when it is determined that the brightness of the environment where the hand is located is greater than a preset upper limit value of brightness, g(x, y) = f(x, y) - 50;

[0061] wherein if g(x, y) is less than 0, g(x, y) = 0.

[0062] In this way, the gray scale of each pixel in the gray scale image can be adjusted according to the brightness of the environment where the hand is located, so as to effectively adjust the gray scale of the image to be processed, thereby facilitating the improvement of the result of subsequent article identification processing.

[0063] Optionally, in the embodiment of the application, the processor is specifically configured to:

[0064] When the brightness of the environment where the hand is located is less than a preset lower limit value of brightness or greater than a preset upper limit value of brightness, it is determined that the image to be processed needs to be adjusted in grayscale.

[0065] In this way, whether the grayscale of the image to be processed needs to be adjusted can be determined according to the brightness of the environment where the hand is located, so that the influence of brightness on the identification of the object in the image can be eliminated, and the accuracy of the identification of the object can be improved.

[0066] Specifically, in the embodiment of the present application, when the brightness of the environment where the hand is located is determined, the following methods can be included:

[0067] Method 1: determined by using a brightness detector.

[0068] Therefore, optionally, in the embodiment of the present application, as shown in Figure 2 , it further includes:

[0069] The brightness detector 103 is configured to detect the brightness of the environment where the hand is located when the door of the smart refrigerator is opened.

[0070] In this way, the detection of the brightness of the environment can be realized by the brightness detector, so that the brightness of the environment where the hand is located can be directly, quickly and effectively determined, thereby facilitating the determination of whether the grayscale of the image to be processed needs to be adjusted.

[0071] Specifically, in the embodiment of the present application, the brightness detector can be a light intensity illuminometer.

[0072] The light intensity illuminometer (also known as a luxmeter) is an instrument specially designed to measure illuminance, which is used to measure the degree of illumination (or brightness). It is usually composed of a selenium photocell or a silicon photocell, a filter and a microammeter. When light shines on the surface of the cell, a photoelectric effect can be generated at the interface, and the size of the photo-generated current is proportional to the illuminance on the surface of the cell.

[0073] If the cell is connected to an external circuit, there will be a current (i.e. photo-generated current) passing through, and the size of the photo-generated current depends on the intensity of the incident light, and the current value can be expressed in units of lux to represent the size of the light intensity illuminance.

[0074] As shown in Figure 2 , if the light intensity illuminometer is set at a position, when the detected light intensity illuminance is 1150Lux, the image of the seafood fungus is accessed (as shown in the left image of Figure 3 ), and when the light intensity illuminance is 140Lux, the image of the cherry tomato is accessed (as shown in the right image of Figure 3 ), experiments have proved that it is difficult to accurately identify the food material in the case of too bright or too dark light intensity illuminance.

[0075] Of course, the brightness detector can be other devices that can realize brightness detection, and is not limited herein.

[0076] Specifically, in the embodiment of the present application, as shown in Figure 2 the brightness detector 103 is arranged in the middle region of the refrigerating chamber of the smart refrigerator.

[0077] The image collector 101 is arranged at the top end of the smart refrigerator.

[0078] As shown in Figure 2 , 104 represents the illuminating lamp, and when the brightness detector has a sensing surface, the sensing surface needs to face the top end of the smart refrigerator, so as to be exposed to the illuminating lamp 104 at the top of the cabinet of the smart refrigerator and other light sources in the room, so as to facilitate more convenient and effective detection of brightness.

[0079] For example, as shown in Figure 2 , the brightness detector 103 is arranged directly above the drawer of the refrigerating chamber (as shown by the dashed box 1 in Figure 2 ).

[0080] At the same time, arranging the image collector at the top end of the smart refrigerator can effectively capture the image to be processed, so as to dynamically identify the articles when the user accesses the articles, and as much as possible to avoid the occlusion between the articles, and at the same time, can guide or remind based on the identification result during the user accessing the articles, such as recommending the storage position and storage mode of the articles.

[0081] Optionally, the processor can be arranged together with the image collector to constitute a collecting module, when the user opens the door of the smart refrigerator, the light of the smart refrigerator is turned on, the collecting module is popped out, the image collector captures and collects images at regular time and sends to the processor, the processor detects the received images to find out the image including hand information, and then determines the image including hand information as the image to be processed for subsequent gray scale adjustment and article identification processing. When the door of the smart refrigerator is closed, the collecting module is retracted, and at this time the image collector no longer collects images.

[0082] In this way, only when the door of the smart refrigerator is opened to collect images, not only can reduce the power consumption, but also can reduce the operation amount of the processor and reduce the manufacturing cost of the processor.

[0083] Of course, the processor can also not be arranged together with the image collector, that is, only the image collector constitutes the collecting module; specifically, only the image collector is popped out and collects images at regular time when the smart refrigerator is opened, and the image collector is retracted when the door of the smart refrigerator is closed.

[0084] In this way, the structure of the acquisition module can be simplified, the volume of the acquisition module can be reduced, and the space occupied by the top of the smart refrigerator can be reduced, which is conducive to optimizing the volume of the smart refrigerator.

[0085] Method 2: Determine using the grayscale of the image to be processed.

[0086] Optionally, in this embodiment of the present invention, the processor is specifically configured to:

[0087] Determine the average grayscale of the grayscale image corresponding to the image to be processed;

[0088] When the average grayscale is less than a preset first grayscale, determining that the brightness of the environment where the hand is located is less than a lower brightness limit;

[0089] When the average grayscale is greater than a preset second grayscale, it is determined that the brightness of the environment in which the hand is located is greater than a preset upper brightness limit; wherein the second grayscale is greater than the first grayscale.

[0090] The first grayscale and the second grayscale can be set according to actual needs and are not limited here.

[0091] For example, when the grayscale range is 0-255, assuming the first grayscale is 50 and the second grayscale is 200, then:

[0092] If the average grayscale is 34.375 (the corresponding image to be processed is Figure 4 As shown in the left figure in the figure, since 34.375 is less than 50, it is determined that the brightness of the environment where the hand is located is less than the lower brightness limit;

[0093] If the average grayscale is 207.237 (the corresponding image to be processed is Figure 4 ), since 207.237 is greater than 200, it is determined that the brightness of the environment where the hand is located is greater than the upper brightness limit.

[0094] In this way, after analyzing the grayscale of the image to be processed, the brightness of the hand's environment can be determined based on the average grayscale obtained from the analysis. Specifically, a larger average grayscale indicates a higher brightness of the hand's environment, while a smaller average grayscale indicates a lower brightness of the hand's environment. Therefore, based on the grayscale of the image to be processed, the brightness of the hand's environment can also be determined, thereby determining whether the grayscale of the image to be processed needs to be adjusted.

[0095] Moreover, this method does not require additional detection devices and can rely solely on existing devices, thereby helping to reduce the production cost of the smart refrigerator.

[0096] In summary, in specific implementation, when determining the brightness of the environment in which the hand is located, either of the above-mentioned methods 1 and 2 can be adopted, and can be selected and set according to actual conditions, and is not limited here.

[0097] Optionally, in the embodiments of the present application, when determining the image to be processed including hand information, the processor can adopt the following process:

[0098] When the image collector collects images in time and sends them to the processor, the processor detects each received image according to a preset hand detection model to detect which or which image includes hand information, and determines the image including hand information as the image to be processed.

[0099] The preset hand detection model can be any model known to those skilled in the art that can realize hand information detection, such as but not limited to machine learning or deep neural network training classification model, which is not limited here.

[0100] Optionally, in the embodiments of the present application, when the processor performs the article identification processing on the image to be processed after adjusting the gray scale, the following method can be adopted:

[0101] The traditional feature extractor and classifier are used to realize feature extraction, classifier learning and training, and test identification of the article image, wherein the feature extractor includes HOG (Histogram of Oriented Gradient), LBP (Local Binary Pattern), DPM (Deformable Part Model), etc., and the classifier can also use SVM (Support Vector Machines), Adaboost, decision tree, Bayesian network, neural network, etc.

[0102] Or, a deep learning method is used to train and identify the article category, including Faster R-CNN, YOLO series algorithm, SSD, etc.

[0103] Through the deep learning method, the article identification can be performed in the process of accessing the article to determine the article category information; and then:

[0104] When the intelligent refrigerator includes a player, the recognition result can be announced through the player;

[0105] And / or, when the intelligent refrigerator includes a display, the recognition result can be displayed through the display.

[0106] Based on the same inventive concept, the embodiments of the present application provide an article identification method, which is similar to the implementation principle of the foregoing intelligent refrigerator, and the specific implementation of the method can be referred to the specific embodiments of the foregoing intelligent refrigerator, and the repeated parts will not be described here.

[0107] Specifically, the embodiment of the present application provides an article identification method, which comprises Figure 5 as shown, can comprise:

[0108] S501, determining the collected image comprising hand information as a to-be-processed image;

[0109] S502, when it is determined that the to-be-processed image needs to adjust the gray scale, adjusting the gray scale of the to-be-processed image according to the brightness of the environment where the hand is located;

[0110] S503, performing article identification processing on the to-be-processed image after adjusting the gray scale.

[0111] Wherein, when it is determined that the to-be-processed image does not need to adjust the gray scale, the to-be-processed image can be directly subjected to article identification processing.

[0112] The article identification process provided by the embodiment of the present application will be described below in combination with specific embodiments.

[0113] Embodiment: taking the smart refrigerator comprising a brightness detector as an example for description, and each of the following steps is executed by a processor.

[0114] Step 1, when the door of the smart refrigerator is opened, receiving the image collected by the image collector in a regular manner, and analyzing the received image to determine the image comprising hand information, and defining the determined image as a to-be-processed image;

[0115] Wherein, the image collector can be but not limited to a camera.

[0116] Step 2, receiving the brightness of the environment where the hand is located detected by the brightness detector, and determining whether the received brightness is less than the lower limit value of brightness; if yes, executing step 3; if no, executing step 6;

[0117] Step 3, converting the to-be-processed image into a gray scale image, and taking the sum of the gray scale of each pixel in the gray scale image and a preset first constant as the adjusted gray scale of the pixel, and when the adjusted gray scale is greater than the preset upper limit value of the gray scale, taking the upper limit value of the gray scale as the adjusted gray scale;

[0118] Step 4, performing gamma correction processing on the adjusted gray scales;

[0119] Step 5, performing article identification processing on the processed to-be-processed image; ending the process;

[0120] Step 6, determining whether the received brightness is greater than the upper limit value of brightness; if yes, executing step 7; if no, returning to step 5;

[0121] Step 7, converting the image to be processed into a gray scale image, and taking the difference between the gray scale of each pixel in the gray scale image and a preset second constant as the adjusted gray scale of the pixel, and taking the lower limit of the gray scale as the adjusted gray scale when the adjusted gray scale is less than the preset lower limit of the gray scale; returning to step 4.

[0122] Embodiment: Taking the smart refrigerator without a brightness detector as an example for illustration, and each of the following steps is completed by the processor.

[0123] Step 1, receiving the image collected by the image collector in a timely manner when the door of the smart refrigerator is opened, and analyzing the received image to determine the image including hand information, and defining the determined image as an image to be processed;

[0124] Step 2, converting the image to be processed into a gray scale image, and determining the average gray scale of the gray scale image;

[0125] Step 3, determining whether the average gray scale is less than the first gray scale; if yes, executing step 4; if no, executing step 7;

[0126] Step 4, taking the sum of the gray scale of each pixel in the gray scale image and a preset first constant as the adjusted gray scale of the pixel, and taking the upper limit of the gray scale as the adjusted gray scale when the adjusted gray scale is greater than the preset upper limit of the gray scale;

[0127] Step 5, performing gamma correction processing on the adjusted gray scales;

[0128] Step 6, performing item recognition processing on the processed image to be processed; ending the process;

[0129] Step 7, determining whether the average gray scale is greater than the second gray scale; if yes, executing step 8; if no, returning to step 6;

[0130] Step 8, taking the difference between the gray scale of each pixel in the gray scale image and a preset second constant as the adjusted gray scale of the pixel, and taking the lower limit of the gray scale as the adjusted gray scale when the adjusted gray scale is less than the preset lower limit of the gray scale; returning to step 5.

[0131] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A smart refrigerator, characterized in that: include: Image collector; The processor is configured to: determining the image captured by the image collector and including the hand information as an image to be processed; When it is determined that the grayscale of the image to be processed needs to be adjusted, the grayscale of the image to be processed is adjusted according to the brightness of the environment in which the hand is located; the processor is specifically configured to: determine a grayscale image corresponding to the image to be processed; When it is determined that the brightness of the environment in which the hand is located is less than a preset brightness lower limit, the sum of the grayscale of each pixel in the grayscale image and a preset first constant is used as the adjusted grayscale of the pixel, and when the adjusted grayscale is greater than a preset grayscale upper limit, the grayscale upper limit is used as the adjusted grayscale; When it is determined that the brightness of the environment in which the hand is located is greater than a preset upper brightness limit, the difference between the grayscale of each pixel in the grayscale image and a preset second constant is used as the adjusted grayscale of the pixel, and when the adjusted grayscale is less than a preset lower grayscale limit, the lower grayscale limit is used as the adjusted grayscale; performing gamma correction processing on each of the adjusted grayscales; During the gamma correction process, the gamma value satisfies the following conditions: the gamma value used when determining that the brightness of the environment in which the hand is located is less than a preset lower brightness limit is different from the gamma value used when determining that the brightness of the environment in which the hand is located is greater than a preset upper brightness limit; wherein, the brightness of the environment in which the hand is located is determined by using a brightness detector or by using the grayscale of the image to be processed; The object recognition process is performed on the image to be processed after the grayscale is adjusted.

2. The smart refrigerator according to claim 1, wherein: The processor is specifically configured to: Normalizing the adjusted grayscales to obtain real values ​​corresponding to the grayscales; Performing the gamma correction process on each of the real values ​​according to a preset gamma value and a gamma correction formula; Convert each of the real values ​​after the gamma correction process into grayscale.

3. The smart refrigerator according to claim 2, wherein: The processor is specifically configured to: The gamma correction process is performed on each of the real values ​​using the following formula: k(x,y)=h(x,y) 1 / gamma ; Wherein, k(x,y) represents the gamma-corrected value corresponding to the pixel located at the xth row and yth column, h(x,y) represents the real value corresponding to the grayscale of the pixel located at the xth row and yth column, and gamma represents the gamma value.

4. The smart refrigerator according to claim 1, wherein: The processor is specifically configured to: When the brightness of the environment in which the hand is located is less than a preset lower brightness limit or greater than a preset upper brightness limit, it is determined that the grayscale of the image to be processed needs to be adjusted.

5. The smart refrigerator according to claim 4, wherein: Also includes: The brightness detector is configured to detect the brightness of the environment in which the hand is located when the door of the smart refrigerator is opened.

6. The smart refrigerator according to claim 5, wherein: The brightness detector is arranged in the middle area of ​​the refrigerator compartment of the smart refrigerator; The image collector is arranged on the top of the smart refrigerator.

7. The smart refrigerator according to claim 4, wherein: The processor is specifically configured to: Determine the average grayscale of the grayscale image corresponding to the image to be processed; When the average grayscale is less than a preset first grayscale, determining that the brightness of the environment in which the hand is located is less than the brightness lower limit; When the average grayscale is greater than a preset second grayscale, determining that the brightness of the environment in which the hand is located is greater than a preset upper brightness limit; The second grayscale is greater than the first grayscale.

8. A method for identifying an object, characterized in that: include: Determining the collected image including the hand information as an image to be processed; When it is determined that the grayscale of the image to be processed needs to be adjusted, the grayscale of the image to be processed is adjusted according to the brightness of the environment in which the hand is located; wherein, the grayscale of the image to be processed is adjusted according to the brightness of the environment in which the hand is located, including: determining the grayscale image corresponding to the image to be processed; when it is determined that the brightness of the environment in which the hand is located is less than a preset lower brightness limit, taking the sum of the grayscale of each pixel in the grayscale image and a preset first constant as the grayscale of the pixel after adjustment, and when the adjusted grayscale is greater than a preset grayscale upper limit, taking the grayscale upper limit as the adjusted grayscale; when it is determined that the brightness of the environment in which the hand is located is greater than the preset upper brightness limit, taking the sum of the grayscale of each pixel in the grayscale image and a preset first constant as the grayscale of the pixel after adjustment The difference between the grayscale of a pixel in the grayscale image and a preset second constant is used as the grayscale after adjustment of the pixel, and when the adjusted grayscale is less than a preset grayscale lower limit, the grayscale lower limit is used as the adjusted grayscale; gamma correction processing is performed on each of the adjusted grayscales; wherein, during the gamma correction processing, the gamma value satisfies the following conditions: the gamma value used when determining that the brightness of the environment where the hand is located is less than the preset brightness lower limit is different from the gamma value used when determining that the brightness of the environment where the hand is located is greater than the preset brightness upper limit; wherein, when determining the brightness of the environment where the hand is located, the following methods are included: determining by using a brightness detector, or determining by using the grayscale of the image to be processed; The object recognition process is performed on the image to be processed after the grayscale is adjusted.

Images