Deep learning electronic product appearance defect automatic detection method
Through deep learning technology, the appearance defects of electronic products are automatically identified, which solves the problems of low efficiency, high cost and inconsistent results of traditional manual detection, and achieves efficient and accurate defect detection, reducing labor costs and improving the controllability of the production process.
Patent Information
- Application Number
- CN202411036851.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-12-31
AI Technical Summary
The appearance defect detection of traditional electronic products relies on manual detection, which is inefficient, prone to errors, high cost, and is easily affected by subjective factors, resulting in inconsistent detection results.
Deep learning technology is adopted to collect and label the appearance images of electronic products, perform data augmentation processing, build a convolutional neural network model, and optimize the model using training data sets to automatically identify defect types and locations.
It improves the accuracy and efficiency of testing, reduces labor costs, enhances the controllability of the production process, promptly discovers and deals with problems in the production process, and improves product quality and market competitiveness.
Smart Images
Figure CN119228721A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electronic products, and in particular to a deep learning method for automatically detecting appearance defects of electronic products. Background Art
[0002] With the continuous development of science and technology, electronic products play an increasingly important role in people's daily lives. However, in the production process of electronic products, the appearance defects not only affect the appearance of the product, but are also likely to affect the performance and service life of the product. Traditional appearance defect detection methods mainly rely on manual inspection, but this method is inefficient, error-prone, and costly. In recent years, the rapid development of deep learning technology has provided new possibilities for the automatic detection of appearance defects in electronic products.
[0003] In traditional electronic product appearance defect detection, manual inspection is the mainstream method. However, manual inspection has many limitations, such as low inspection efficiency, increased false detection rate due to fatigue of workers after long hours of work, and high labor costs. In addition, manual inspection is easily affected by the subjective factors of the inspectors, resulting in inconsistent inspection results.
[0004] Deep learning is an important branch of artificial intelligence. It simulates the learning process of the human brain by building a deep neural network model. In the detection of appearance defects of electronic products, deep learning technology can automatically extract features from images and identify appearance defects by training models. This method not only improves the accuracy and efficiency of detection, but also greatly reduces labor costs. Summary of the invention
[0005] The purpose of the present invention is to provide a deep learning method for automatic detection of electronic product appearance defects, which solves the problem that in the prior art, manual detection is the mainstream method in traditional electronic product appearance defect detection. However, manual detection has many limitations, such as low detection efficiency, increased false detection rate due to fatigue of workers after long hours of work, and high labor costs. In addition, manual detection is also easily affected by the subjective factors of the detection personnel, resulting in inconsistent detection results.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] A deep learning method for automatic detection of electronic product appearance defects includes collecting electronic product appearance images and marking them according to different defect types; performing data enhancement processing on the collected images; constructing a suitable convolutional neural network model; using a training data set to train the constructed convolutional neural network model; inputting the electronic product appearance image to be detected into the trained model to perform defect recognition; and outputting the defect type and location information on the electronic product appearance according to the recognition result of the model.
[0008] Preferably, data enhancement processing includes but is not limited to flipping, cropping, scaling and denoising operations to increase the diversity and generalization ability of the data set.
[0009] Preferably, the convolutional neural network model includes multiple convolutional layers, pooling layers and fully connected layers for extracting image features and identifying defects.
[0010] Preferably, an optimization algorithm is used to optimize the model parameters during the model training process to improve the recognition accuracy of the model.
[0011] Preferably, during the defect recognition process, the model can automatically recognize various defects on the appearance of the electronic product and output defect type and location information.
[0012] Preferably, the output defect type and location information can be used to guide quality control and improvement measures during the production process.
[0013] Preferably, it can be applied to appearance defect detection of various types of electronic products, including but not limited to mobile phone screens, computer monitors, tablet computers, etc.
[0014] The present invention has at least the following beneficial effects:
[0015] Improved detection accuracy and efficiency: Deep learning technology is used to automatically extract image features and identify defects, avoiding the problem of false detection and missed detection caused by subjectivity and fatigue factors in manual detection. Reduced labor costs: The automatic detection method of the present invention can replace a large amount of manual detection work, thereby reducing labor costs. Enhanced controllability of the production process: By real-time monitoring and identifying defects on the appearance of electronic products, problems in the production process can be discovered and handled in a timely manner, improving the controllability of the production process. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.
[0017] Figure 1 It is a structural schematic diagram of the steps of automatically detecting appearance defects of electronic products according to the present invention. DETAILED DESCRIPTION
[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0019] Embodiment 1, refer to Figure 1 A deep learning method for automatic detection of electronic product appearance defects includes collecting electronic product appearance images and marking them according to different defect types; performing data enhancement processing on the collected images; constructing a suitable convolutional neural network model; using a training data set to train the constructed convolutional neural network model; inputting the electronic product appearance image to be detected into the trained model to perform defect recognition; and outputting the defect type and location information on the electronic product appearance according to the recognition result of the model.
[0020] According to the above embodiment, the method includes the following steps:
[0021] Dataset collection and labeling: Collect images of the appearance of electronic products and label them according to different defect types to form training data sets and test data sets.
[0022] Data augmentation: Data augmentation is performed on the collected images to increase the diversity and generalization ability of the dataset.
[0023] Construct a convolutional neural network model: Based on the appearance characteristics of electronic products, construct a suitable convolutional neural network model for feature extraction and defect identification.
[0024] Model training: Use the training data set to train the constructed convolutional neural network model, and improve the recognition accuracy of the model by continuously optimizing the model parameters.
[0025] Defect recognition: Input the appearance image of the electronic product to be inspected into the trained model for defect recognition.
[0026] Result output: Based on the recognition results of the model, the defect type and location information on the appearance of the electronic product are output.
[0027] Compared with the prior art, the present invention has the following beneficial effects:
[0028] Improved detection accuracy and efficiency: Deep learning technology is used to automatically extract image features and identify defects, avoiding false detection and missed detection problems caused by subjectivity and fatigue factors in manual detection.
[0029] Reduced labor costs: The automatic detection method of the present invention can replace a large amount of manual detection work, thereby reducing labor costs.
[0030] Enhanced controllability of the production process: By real-time monitoring and identifying defects in the appearance of electronic products, problems in the production process can be discovered and handled in a timely manner, improving the controllability of the production process.
[0031] Improved product quality: By accurately identifying and handling appearance defects, the quality level of electronic products can be improved, enhancing consumers' purchasing intention and user experience.
[0032] Example 2, refer to Figure 1 ,Data augmentation processing includes but is not limited to flipping, cropping, scaling and ,noise reduction operations to increase the diversity and generalization ability of ,the dataset.
[0033] The convolutional neural network model includes multiple convolutional layers, pooling layers, and fully connected layers to extract image features and identify defects.
[0034] During the model training process, an optimization algorithm is used to optimize the model parameters to improve the recognition accuracy of the model.
[0035] During the defect recognition process, the model can automatically identify various defects on the appearance of electronic products and output defect type and location information.
[0036] The output defect type and location information can be used to guide quality control and improvement measures during the production process.
[0037] It can be applied to various types of electronic product appearance defect detection, including but not limited to mobile phone screens, computer monitors, tablet computers, etc.
[0038] Technological advancement:
[0039] By applying deep learning technology, the present invention achieves efficient and accurate identification of appearance defects of electronic products. Compared with traditional manual inspection or simple image processing technology, it has higher technical advancement and intelligence level.
[0040] Improved detection efficiency:
[0041] The automated inspection process greatly improves the inspection speed. The traditional manual inspection process is time-consuming and error-prone, while the automatic inspection method of the present invention can process a large number of products in a short time, significantly improving production efficiency.
[0042] Cost Savings:
[0043] The human resources required for manual inspection are reduced, thereby reducing labor costs. In addition, due to the improvement of inspection efficiency, the product's residence time in the inspection link is reduced, further reducing inventory and operating costs.
[0044] Improved quality control capabilities:
[0045] The present invention can timely discover and identify minor defects in product appearance, which helps enterprises to carry out quality control in the early stages of production and prevent inferior products from entering the market, thereby protecting brand image and consumer trust.
[0046] Production process optimization:
[0047] Real-time feedback of defect information can help the production line adjust process parameters in a timely manner, reduce the production of defective products, and achieve continuous optimization of the production process.
[0048] Enhanced market competitiveness:
[0049] High-quality products and efficient production processes can enhance the market competitiveness of enterprises. Through the present invention, enterprises can quickly respond to market demands and provide products with flawless appearance and stable performance, thus standing out in the fierce market competition.
[0050] Scalability and adaptability:
[0051] The deep learning model of the present invention can adapt to new defect types and electronic product models through continuous learning and training, has good scalability and adaptability, and can meet the needs of continuous enterprise development and market changes.
[0052] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.
Claims
1. A deep learning method for automatic detection of electronic product appearance defects, characterized in that: The following steps are involved: Collect images of electronic product appearance and mark them according to different defect types; Perform data augmentation on the collected images; Build a suitable convolutional neural network model; The constructed convolutional neural network model is trained using the training data set; the appearance image of the electronic product to be inspected is input into the trained model for defect recognition; and the defect type and location information on the appearance of the electronic product is output based on the recognition results of the model.
2. According to claim 1, a deep learning method for automatic detection of electronic product appearance defects is characterized in that: Data augmentation processing includes but is not limited to flipping, cropping, scaling, and denoising operations to increase the diversity and generalization ability of the dataset.
3. The method for automatic detection of electronic product appearance defects based on deep learning according to claim 2 is characterized in that: The convolutional neural network model includes multiple convolutional layers, pooling layers, and fully connected layers to extract image features and identify defects.
4. The method for automatic detection of electronic product appearance defects based on deep learning according to claim 3 is characterized in that: During the model training process, an optimization algorithm is used to optimize the model parameters to improve the recognition accuracy of the model.
5. The method for automatic detection of electronic product appearance defects based on deep learning according to claim 2, characterized in that: During the defect recognition process, the model can automatically identify various defects on the appearance of electronic products and output defect type and location information.
6. The method for automatic detection of electronic product appearance defects based on deep learning according to claim 5, characterized in that: The output defect type and location information can be used to guide quality control and improvement measures during the production process.
7. The method for automatic detection of electronic product appearance defects based on deep learning according to claim 6, characterized in that: It can be applied to various types of electronic product appearance defect detection, including but not limited to mobile phone screens, computer monitors, tablet computers, etc.