How to detect display Mura?
Best insight from top research papers
Display Mura can be detected using various methods. One approach is to use a CNN with multiple sets of convolution and max-pooling layers, which can achieve an accuracy of over 0.8 . To improve the detection of low contrast Mura, a contrast-enhancement method and a subspace method can be employed, resulting in an accuracy close to human visible test . Another method involves using a demoireing framework to remove moire patterns in captured images, allowing for accurate Mura detection . A progressive hybrid model, based on human visual perception, consisting of multiclass CNNs and residual neural networks, can also be used for accurate classification between normal displays and weak Mura . Additionally, a GAN-based model called UADD-GAN, which uses a generator and discriminator, has shown promising results in Mura detection . Finally, a method involving the use of a neural network model trained on labeled samples can effectively detect Mura in display screens .
Answers from top 5 papers
More filters
| Papers (5) | Insight |
|---|---|
| The paper describes a method for detecting Mura of a display screen by collecting an image, sampling it, obtaining a gray curve, extracting a target feature, training a neural network model, and detecting the presence of Mura in a to-be-detected image. | |
| The paper proposes using a Convolutional Neural Network (CNN) with multiple sets of convolution and max-pooling layers to detect display Mura with an accuracy of more than 0.8. Additionally, contrast-enhancement and subspace methods are employed to improve detection of low contrast Mura. | |
3 Citations | The paper proposes a GAN-based model called UADD-GAN to detect Mura in display panels. The model is trained using normal samples and is able to distinguish normal images from those with Mura based on their reconstructions. |
| The paper proposes a new demoireing framework using U-Net to remove moire patterns in captured images, thereby accurately detecting Mura defects in displays. | |
| The paper proposes a method using machine learning, specifically a "Progressive Hybrid model," to automatically detect and classify display Mura in the front-end process of display manufacturing. |
Related Questions
What are the kinetic parameters of MurA?4 answersMurA is an enzyme involved in peptidoglycan biosynthesis. It catalyzes the first committed step in this process. The breakdown of MurA's tetrahedral intermediate (THI) was found to be slower than the normal reaction, indicating slow binding and a conformationally constrained active site. The MurA cocrystal structure revealed a new conformation in which the Arg397 side chain tracked phosphite out of the catalytic site. This conformational change involved breaking several ion pairs and hydrogen bonds, and Cys115 appears to play a role in forming this staged conformation. The kinetic parameters of MurA include a kcat/KM value of 520 M^-1 s^-1 for the breakdown of THI. Additionally, a text mining system was developed to aid in the search for experimentally obtained parameters for kinetic models, which outperformed keyword searches in abstracts by a factor of five in terms of precision.
How to detect LPI/LPD signals?5 answersLPI/LPD signals can be detected using various methods. One approach is to use non-parametric source enumeration methods, such as the elbow-based source enumeration (EBSE) method, which separates the signal and noise eigenvalues using elbow estimation. Another method is to employ interference detection and classification (IDC) algorithms, which can identify and classify different types of interference signals with high accuracy. Additionally, coherent processing techniques can be used for low SNR LPI/LPD communication systems, where precise synchronization in time, frequency, and phase is required. Data-aided joint frame and carrier synchronization algorithms, such as the generalized likelihood ratio test (GLRT), can be used to detect embedded preamble signals and provide carrier estimates for coherent demodulation.
How is AI content detected?5 answersAI content is detected using various methods. One approach is to use text classification models that are trained on linguistic features and one-class learning techniques. These models can accurately detect AI-generated essays even without access to positively labeled data for training. Another method involves embedding watermarks into AI-generated content before release and detecting the presence of similar watermarks in the content to identify AI-generated content. However, this watermark-based detection method has been found to be insufficient as attackers can post-process AI-generated images with imperceptible perturbations to evade detection while maintaining visual quality. Additionally, technologies have been developed to analyze content and determine if it was generated by an artificial intelligence source, displaying the determination accordingly. These different approaches highlight the need for new and improved methods to detect AI-generated content.
How to detect board games?5 answersBoard games can be detected using various methods. One approach is to describe game boards using underlying graphs based on tiling, shape, and graph operators, allowing for simple and succinct descriptions. Another method involves using a board game apparatus with marked paths and subunits, where the numbers on the subunits are divisible by the numbers on the faces of a die. In the field of image recognition, a chessboard recognition method has been developed that identifies the outer frame of a chessboard, calculates the coordinates of intersection points, and accurately positions the chess pieces on the board. Additionally, a novel approach uses edge detection, clustering, and geometric constraints to recognize chessboards, even when populated with chess pieces, achieving high accuracy.
How to detect fake news?3 answersFake news detection involves identifying reports that are intentionally false or hoaxes disseminated through mainstream news sources or social media platforms. Several studies have been conducted to detect fake news using machine learning techniques. Deep learning methods such as CNN, Bidirectional LSTM, and ResNet, combined with pre-trained word embedding, have shown promising results. Fact-checking is also important in detecting fake news. Machine learning models such as Support Vector Machine, Naive Bayes, and Neural Network have been used for fake news detection, with Support Vector Machine achieving high accuracy. Feature extraction and machine learning algorithms, including Long-Short-Term Memory, have been proposed for fake news detection, showing higher performance compared to benchmark models.
Can ai detect canser?2 answersYes, AI can detect cancer. One paper proposes the use of Explainable Artificial Intelligence (XAI) techniques to detect the effectiveness of perturbations crafted by adversarial attacks on a trained model for lung cancer detection. Another paper presents a low-cost technique for effectively diagnosing various faults in a gear bearing system using AI. The technique involves monitoring vibration profiles of different gears and bearings and training a machine learning model on statistical features extracted from these profiles. Additionally, a paper discusses the use of AI-based DDoS defending models to defend against DDoS attacks without manual intervention. However, it highlights the risk of malicious training, known as the Artificial Intelligence Trojan attack, and proposes a GAN-based AI Robustness test algorithm to verify the robustness of the AI model.