3C product automatic detection system and method
By automatically collecting and analyzing the packaging box images and weights of 3C products, combined with machine learning algorithms, realization of authenticity detection without unpacking is achieved, solving the problem of manual confirmation in the existing technology, improving detection efficiency and ensuring the authenticity of the goods sex.
Patent Information
- Application Number
- CN202411037510.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-12-03
AI Technical Summary
During the 3C product storage process, the prior art requires manual unpacking and repackaging to confirm the authenticity of the goods, which is time-consuming and unpopular with customers.
A 3C product automatic detection system is designed, including product information collection module, weighing module, control module, early warning module and execution device. By automatically collecting image information and actual weight on the outer side of the packaging box, the authenticity of the product is identified by machine learning algorithms. Implement exception detection and partition processing without unpacking.
It realizes automated 3C product inspection, accurately identify packaging and internal structures, ensures the authenticity of goods, improves inspection efficiency and reduces the need for manual operation.
Smart Images

Figure CN119063784A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of product detection, and in particular to a 3C product automatic detection system and method. Background Art
[0002] "3C products" are the combination of computers, communications and consumer electronics, also known as "information appliances". Since 3C products are generally not large in size, the word "small" is often added in the middle, so they are often collectively referred to as "3C small appliances". 3C products usually refer to computers, tablet computers, mobile phones, digital cameras, walkmans, electronic dictionaries, audio and video playback hardware devices or digital audio players, etc.
[0003] In the warehousing process of 3C products, after receiving the 3C products, the warehouse will manually check the mobile phone packaging and record the serial number. To ensure the authenticity of the goods, the packaging needs to be opened for identification and confirmation, and then re-sealed. This post-sealing machine is difficult to be accepted by all customers, and the operation process is very time-consuming. Summary of the invention
[0004] In view of the defects in the prior art, the present invention provides a 3C product automatic detection system and method to solve the technical problems existing in the prior art.
[0005] In a first aspect, an embodiment of the present invention provides a 3C product automatic detection system, including: a product information collection module, a weighing module, a control module, an early warning module, an execution device and an internal structure collection module;
[0006] The product information acquisition module is used to acquire image information of the outer side of the 3C product packaging box to be inspected;
[0007] The weighing module is used to collect the actual weight of the 3C product to be tested;
[0008] The control module is used to determine whether the image information is useful, and if it is useful, pre-process the image information to obtain pre-processed image information, confirm whether the product information and packaging are abnormal according to the pre-processed image information, and compare the actual weight with the product standard to determine whether the product weight is abnormal. If it is abnormal, send an early warning control instruction to the early warning module, and send an instruction to the execution device to transfer the 3C product to the abnormal area;
[0009] The warning module is used to send out a warning signal when there is an abnormality in the 3C product;
[0010] The execution device is used to transfer the product to the abnormal area when there is an abnormality in the 3C product;
[0011] The internal structure acquisition module is used to collect product image information in a 3C product packaging box with a normal weight box outer packaging, and send the product image information to the control module for processing;
[0012] The control module is used to pre-process the product image information and extract image features, input the image features into the trained internal structure recognition model for recognition, obtain the recognition result, the recognition result includes whether the 3C product is a genuine product or a counterfeit product, and send a control instruction to the execution device according to the recognition result;
[0013] The execution device is also used to transmit the real product to the normal area and transmit the counterfeit product to the abnormal area.
[0014] In a second aspect, an embodiment of the present invention provides a 3C product automatic detection method, comprising the following steps: a product information acquisition module acquires image information of the outer side of a 3C product packaging box to be detected, and transmits the image information to a control module;
[0015] The weighing module collects the actual weight of the 3C product to be tested and transmits the actual weight to the control module;
[0016] The control module receives the image information and the actual weight, pre-processes the image information to obtain pre-processed image information, confirms whether the product information and packaging are abnormal according to the pre-processed image information, and compares the actual weight with the product standard to determine whether the product weight is abnormal. If abnormal, an early warning control instruction is sent to the early warning module, and an instruction to transfer the 3C product to the abnormal area is sent to the execution device;
[0017] The early warning module sends out early warning signals when there are abnormalities in 3C products;
[0018] The execution device transfers the product to the abnormal area when there is an abnormality in the 3C product;
[0019] The internal structure acquisition module acquires product image information in a 3C product packaging box with a normal weight box outer packaging, and sends the product image information to the control module for processing;
[0020] The control module receives product image information, pre-processes the product image information, extracts image features, inputs the image features into the trained internal structure recognition model for recognition, obtains recognition results, including whether the 3C product is a genuine product or a counterfeit product, and sends control instructions to the execution device according to the recognition results;
[0021] The execution device transfers the genuine products to the normal area and the counterfeit products to the abnormal area.
[0022] Beneficial effects of the present invention:
[0023] An automatic 3C product detection system and method provided by an embodiment of the present invention can automatically collect external image information of a 3C product packaging box, identify key information and determine whether the packaging is abnormal through the external image information of the packaging box, and automatically weigh the product to determine whether the actual weight of the product is consistent with the standard weight of the product. If it is abnormal or inconsistent, it is transmitted to an abnormal area. In addition, the system and method can also collect and identify images of the internal structure of the 3C product, and use a machine learning algorithm to check whether the product is a real product or a counterfeit product, thereby accurately detecting the key information and internal structure of the 3C product without unpacking the product packaging, thereby ensuring the authenticity of the goods. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.
[0025] Figure 1 A schematic diagram showing the structure of a 3C product automatic detection system provided by the first embodiment of the present invention is shown;
[0026] Figure 2 A flow chart of a 3C product automatic detection method provided by another embodiment of the present invention is shown. DETAILED DESCRIPTION
[0027] 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 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.
[0028] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.
[0029] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.
[0030] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0031] As used in this specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.
[0032] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.
[0033] like Figure 1 As shown, a schematic diagram of the structure of a 3C product automatic detection system provided by the first embodiment of the present invention is shown, the system includes: a product information acquisition module, a weighing module, a control module, an early warning module, an execution device and an internal structure acquisition module; the product information acquisition module is used to collect image information of the outer side of the 3C product packaging box to be detected; the weighing module is used to collect the actual weight of the 3C product to be detected; the control module is used to determine whether the image information is useful, and if the image information is useful, the image information is pre-processed to obtain pre-processed image information, and whether the product information and packaging are abnormal according to the pre-processed image information, and the actual weight is compared with the product standard to determine whether the product weight is abnormal. If it is abnormal, a warning control instruction is sent to the early warning module, and An instruction to transfer the 3C product to the abnormal area is sent to the execution device; the early warning module is used to send an early warning signal when the 3C product is abnormal; the execution device is used to transfer the product to the abnormal area when the 3C product is abnormal; the internal structure acquisition module is used to collect product image information in the 3C product packaging box with normal weight and outer packaging, and send the product image information to the control module for processing; the control module is used to pre-process the product image information and extract image features, input the image features into the trained internal structure recognition model for recognition, and obtain the recognition result, the recognition result includes whether the 3C product is a real product or a counterfeit product, and send a control instruction to the execution device according to the recognition result; the execution device is also used to transfer the real product to the normal area and the counterfeit product to the abnormal area.
[0034] In this embodiment, the execution device is a crawler conveyor for conveying 3C products, and the product information includes serial code, model, memory and other information. The staff puts the 3C products to be tested on the unloading table, and transmits the 3C products to be tested to the 3C product automatic detection system through the crawler conveyor for detection. First, the product information acquisition module automatically collects the image information of the outside of the packaging box of the 3C product to be tested. The product information acquisition module is a scanner or an infrared camera, and the collected image has high clarity and a wide capture range. The product information acquisition module sends the collected image information to the control module for processing. The control module receives the image information and identifies the key information on the packaging box through picture learning and OCR recognition methods. It determines whether the image information is useful based on whether the key information can be identified. According to different products, the three basic views of the main view, top view and left view can be collected for observation. When it is useful, the image information is preprocessed to obtain the preprocessed image information, and the product information and packaging are confirmed to be abnormal based on the preprocessed image information. Then, through the weighing module, the system can preset the standard weight of various 3C products, accurate to 1g (the accuracy can be freely set). The weighing module can use a force sensor, a vibration sensor, a balance, an electronic scale, a platform scale or a spring scale to weigh. The weighing module collects the actual weight of the 3C product to be tested. The control module compares the actual weight with the standard weight of the product to determine whether the product weight meets the standard weight of the product. If it does not meet the standard weight, it indicates an abnormality. If it meets the standard weight, it indicates normal. When the control module finds an abnormal situation, it sends an early warning control instruction to the early warning module and sends an instruction to the execution device to transfer the 3C product to the abnormal area. After receiving the early warning control instruction sent by the control module, the early warning module sends a warning signal. After receiving the instruction sent by the control module to transfer the 3C product to the abnormal area, the execution device transfers the 3C product to the abnormal area.
[0035] When the weight and outer packaging of the 3C product are normal, the execution device continues to transmit the 3C product to the internal structure acquisition module. The internal structure acquisition module collects the product image information in the 3C product packaging box with normal weight and outer packaging, and sends the product image information to the control module. The control module preprocesses the product image information and extracts image features, inputs the image features into the trained internal structure recognition model for recognition, and obtains whether the 3C product is a real product or a counterfeit product. When the 3C product is a real product, a control instruction to transmit the real product to the normal area is sent to the execution device. When the 3C product is a counterfeit product, a control instruction to transmit the counterfeit product to the abnormal area is sent to the execution device. The execution device executes the control instruction sent by the control module, transmits the real product to the normal area, and transmits the counterfeit product to the abnormal area. The products transmitted to the abnormal area wait for manual verification.
[0036] In another embodiment of the present invention, considering that the position or posture of the product may change when it is transmitted on the crawler, an adjustment module is set to adjust the position or posture of the 3C product in another embodiment of the present invention. The adjustment module can use a robotic arm. When the first acquisition fails, when the control module finds that the acquired image information is useless, the control module will control the early warning module to send an early warning signal, and control the robotic arm to automatically adjust, and complete multi-angle information acquisition by flipping the plane, etc., to improve the integrity of product information acquisition and ensure the accuracy of later product information recognition.
[0037] In another embodiment of the present invention, the control module includes a preprocessing unit, which is used to perform denoising, image enhancement and edge detection on the image information. By preprocessing the image information by the preprocessing unit, the weight of subsequent information extraction can be increased and the accuracy of subsequent detection can be improved.
[0038] The control module also includes a feature extraction unit, which is used to extract image feature information from the processed image information, and the image feature information includes shape features, texture features and color features.
[0039] The feature extraction unit extracts texture features from the processed image information using a wavelet feature extraction method, which can more effectively depict the characteristics of texture according to human vision.
[0040] A two-dimensional Gabor wavelet basis function can be defined as
[0041]
[0042] Where: g(x,y) is the Gaussian function modulated by a complex sine function, σ x ,σ y are the variances of the Gabor wavelet basis function along the x-axis and y-axis, respectively, and ω is the center frequency. Here, the Gabor wavelet basis function is a bandpass filter with (ω,0) as the center frequency. The Gabor wavelet family is formed by scaling and rotating the Gabor wavelet basis function g(x,y):
[0043] g m (x, y) = a -m g(x′y′), a>1, m, n∈z (2)
[0044] Where: x′=a -m (xcosθ+ysinθ),y′=a -m (-cosθ+ysinθ), θ=nπ / k, k is the direction number, a -m is the scale factor.
[0045] In this embodiment, Gabor filters with 5 scales and 4 directions (0°, 45°, 90°, 135°) are used to convolve the input texture image with the Gabor wavelet to obtain subbands of different directions and scales. Assume that the input texture image is I(x, y), then the image output by Gabor decomposition is:
[0046] I mn (x, y) = I (x, y) × g mn (x, y) (3)
[0047] set up. I mn( x,y) is of size M×N, and the mean and variance of its absolute value are calculated to form a 1×40-dimensional feature vector V=[μ mn ,σ mn ]. Let the size be M×N, calculate the mean and variance of its absolute value to form a 1×40-dimensional feature vector V=[μ mn ,σ mn ].
[0048]
[0049] In another embodiment of the present invention, the control module further includes a model building unit, which is used to use a machine learning method to establish an internal structure recognition model based on the labeled real 3C product and counterfeit 3C product labels. The machine learning method can use a support vector machine or a decision tree, and this embodiment uses a support vector machine.
[0050] The internal structure acquisition module uses X-ray imaging technology, ultrasonic imaging technology, electromagnetic imaging technology, neutron imaging technology, Gamma ray imaging technology, lidar imaging technology, infrared imaging technology or millimeter wave imaging technology to collect 3C product image information.
[0051] Specifically, a certain amount of 3C product image information is collected as a data set, including real 3C products and counterfeit 3C product image information, and these data sets are marked. Image features are extracted from product image information, and the feature values and corresponding labels obtained from the image information are used as input values, input into the support vector machine, and the appropriate kernel function is selected to complete training and classification. Use laser scanning equipment to scan the target object to obtain the point cloud data of its internal structure, and ensure that the collected data covers a variety of objects with internal structures so that the model can learn enough diversity and extract useful features, such as shape, size, texture, etc. These features should be able to characterize the internal structure of the object. Select the support vector machine according to the characteristics (such as classification, regression, segmentation, etc.) and data characteristics, and divide the data set: divide the preprocessed data set into training set, validation set and test set, including hyperparameters such as learning rate, batch size, and training rounds. Then we start training the model, use the training set data to train the model, and adjust the model parameters through the back propagation algorithm to minimize the loss function. During the training process, we use the validation set to monitor the performance of the model and make adjustments as appropriate. Then we deploy the trained model to actual application environments, such as robots and automated equipment. We use laser scanning equipment to obtain the internal structure data of the object in real time, and use the model to detect and classify it.
[0052] An automatic 3C product detection system provided by an embodiment of the present invention can automatically collect external image information of 3C product packaging boxes, identify key information and determine whether the packaging is abnormal through the external image information of the packaging boxes, and automatically weigh the product to determine whether the actual weight of the product is consistent with the standard weight of the product. If it is abnormal or inconsistent, it is transmitted to the abnormal area. It can also collect and identify images of the internal structure of the 3C product, and use a machine learning algorithm to check whether the product is a real product or a counterfeit product. It realizes accurate detection of key information and internal structure of 3C products without unpacking the product packaging, thereby ensuring the authenticity of the goods.
[0053] In the above first embodiment, a 3C product automatic detection system is provided. Correspondingly, the present application also provides a 3C product automatic detection method. Figure 2 , which is a flow chart of a 3C product automatic detection method provided by the second embodiment of the present invention. Since the method embodiment is basically similar to the device embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the device embodiment. The method embodiment described below is only illustrative.
[0054] like Figure 2 FIG. 1 is a flowchart of a 3C product automatic detection method provided by another embodiment of the present invention, the method comprising the following steps:
[0055] The product information acquisition module acquires image information of the outer side of the 3C product packaging box to be inspected, and transmits the image information to the control module;
[0056] The weighing module collects the actual weight of the 3C product to be tested and transmits the actual weight to the control module;
[0057] The control module receives the image information and the actual weight, and determines whether the image information is useful. If the image information is useful, the image information is preprocessed to obtain preprocessed image information. The product information and packaging are confirmed to be abnormal according to the preprocessed image information, and the actual weight is compared with the product standard to determine whether the product weight is abnormal. If it is abnormal, an early warning control instruction is sent to the early warning module, and an instruction to transfer the 3C product to the abnormal area is sent to the execution device;
[0058] The early warning module sends out an early warning signal when there is an abnormality in the 3C product, and the execution device transfers the product to the abnormal area when there is an abnormality in the 3C product;
[0059] The internal structure acquisition module acquires the weight and product image information in the 3C product packaging box with normal outer packaging, and sends the product image information to the control module for processing;
[0060] The control module receives product image information, pre-processes the product image information, extracts image features, inputs the image features into the trained internal structure recognition model for recognition, obtains recognition results, including whether the 3C product is a genuine product or a counterfeit product, and sends control instructions to the execution device according to the recognition results;
[0061] The execution device transfers the genuine products to the normal area and the counterfeit products to the abnormal area.
[0062] When the control module determines that the image information is useless, it sends a control instruction for adjusting the position or posture of the 3C product to the adjustment module.
[0063] An automatic detection method for 3C products provided in an embodiment of the present invention can automatically collect external image information of 3C product packaging boxes, identify key information and determine whether the packaging is abnormal through the external image information of the packaging boxes, and automatically weigh the product to determine whether the actual weight of the product is consistent with the standard weight of the product. If it is abnormal or inconsistent, it is transmitted to the abnormal area. In addition, the internal structure image of the 3C product can be collected and identified, and a machine learning algorithm is used to check whether the product is a real product or a counterfeit product. The key information and internal structure of the 3C product can be accurately detected without unpacking the product packaging, thereby ensuring the authenticity of the goods.
[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.
Claims
1. A 3C product automatic detection system, characterized in that: include: Product information collection module, weighing module, control module, early warning module, execution device and internal structure collection module; The product information acquisition module is used to acquire image information of the outer side of the 3C product packaging box to be inspected; The weighing module is used to collect the actual weight of the 3C product to be tested; The control module is used to determine whether the image information is useful, and if it is useful, pre-process the image information to obtain pre-processed image information, confirm whether the product information and packaging are abnormal according to the pre-processed image information, and compare the actual weight with the product standard to determine whether the product weight is abnormal. If it is abnormal, send an early warning control instruction to the early warning module, and send an instruction to the execution device to transfer the 3C product to the abnormal area; The warning module is used to send out a warning signal when there is an abnormality in the 3C product; The execution device is used to transfer the product to the abnormal area when there is an abnormality in the 3C product; The internal structure acquisition module is used to collect product image information in a 3C product packaging box with normal weight and outer packaging, and send the product image information to the control module for processing; The control module is used to pre-process the product image information and extract image features, input the image features into the trained internal structure recognition model for recognition, obtain the recognition result, the recognition result includes whether the 3C product is a genuine product or a counterfeit product, and send a control instruction to the execution device according to the recognition result; The execution device is also used to transmit the real product to the normal area and transmit the counterfeit product to the abnormal area.
2. The 3C product automatic detection system according to claim 1, characterized in that: It also includes an adjustment module, which is used to adjust the position or posture of the 3C product when the collected image information is useless.
3. The 3C product automatic detection system according to claim 1, characterized in that: The control module comprises a preprocessing unit, and the preprocessing unit is used for performing denoising, image enhancement and edge detection on image information.
4. The 3C product automatic detection system according to claim 3, characterized in that: The control module further comprises a feature extraction unit, which is used to extract image feature information from the processed image information, wherein the image feature information comprises shape features, texture features and color features.
5. The 3C product automatic detection system according to claim 4, characterized in that: The control module also includes a model building unit, which is used to use a machine learning method to establish an internal structure recognition model based on the marked real 3C product and counterfeit 3C product labels.
6. The 3C product automatic detection system according to claim 5, characterized in that: The machine learning method is a support vector machine or a decision tree.
7. The 3C product automatic detection system according to claim 1, characterized in that: The product information acquisition module is a scanner or an infrared camera.
8. The 3C product automatic detection system according to claim 1, characterized in that: The internal structure acquisition module uses X-ray imaging technology, ultrasonic imaging technology, electromagnetic imaging technology, neutron imaging technology, Gamma ray imaging technology, lidar imaging technology, infrared imaging technology or millimeter wave imaging technology to collect 3C product image information.
9. A 3C product automatic detection method, characterized in that: The method comprises the following steps: The product information acquisition module acquires image information of the outer side of the 3C product packaging box to be inspected, and transmits the image information to the control module; The weighing module collects the actual weight of the 3C product to be tested and transmits the actual weight to the control module; The control module receives the image information and the actual weight, and determines whether the image information is useful. If the image information is useful, the image information is preprocessed to obtain preprocessed image information, and the product information and packaging are confirmed to be abnormal according to the preprocessed image information, and the actual weight is compared with the product standard to determine whether the product weight is abnormal. If it is abnormal, an early warning control instruction is sent to the early warning module, and an instruction to transfer the 3C product to the abnormal area is sent to the execution device; The early warning module sends out early warning signals when there are abnormalities in 3C products; The execution device transfers the product to the abnormal area when there is an abnormality in the 3C product; The internal structure acquisition module acquires the weight and product image information in the 3C product packaging box with normal outer packaging, and sends the product image information to the control module for processing; The control module receives product image information, pre-processes the product image information, extracts image features, inputs the image features into the trained internal structure recognition model for recognition, obtains recognition results, including whether the 3C product is a genuine product or a counterfeit product, and sends control instructions to the execution device according to the recognition results; The execution device transfers the genuine products to the normal area and the counterfeit products to the abnormal area.
10. The 3C product automatic detection method according to claim 9, characterized in that: When the control module determines that the image information is useless, it sends a control instruction for adjusting the position or posture of the 3C product to the adjustment module.
