The implementation of China's higher education talent quality enhancement project integrates modern information technology, particularly digital imageprocessing and virtual simulation technology, to promote the c...
详细信息
ISBN:
(纸本)9798350375343;9798350375336
The implementation of China's higher education talent quality enhancement project integrates modern information technology, particularly digital imageprocessing and virtual simulation technology, to promote the cultivation of talent in universities under the "Intelligent + Education" model. This paper introduces a virtual simulation experimental system for the "Digital imageprocessing" course based on WebGL and Matlab, which centers around a 3D environment and supports online coding, real-time computation, and virtual display, facilitating task-driven online experimental learning. The system, built with a frontend using Vue, ***, and *** in conjunction with a MySQL database, enables user interaction with the 3D environment, enhancing the real-time and interactive nature of the experiments. It transforms traditional teaching methods and encourages students to engage in active and in-depth learning. The trial run of the system has demonstrated good compatibility, stability, and cross-platform capabilities.
Recently, providing real-time navigation of unmanned aerial vehicles independent of global positioning systems has become of great importance. The state-of-the-art methods based on deep learning, which give good resul...
详细信息
ISBN:
(纸本)9798350388978;9798350388961
Recently, providing real-time navigation of unmanned aerial vehicles independent of global positioning systems has become of great importance. The state-of-the-art methods based on deep learning, which give good results in certain datasets, and the existing methods can not provide real-time and good solutions on images with dynamic and fast moving. Moreover, the methods, were developed so far, were focused on object-based tracking algorithms. In this paper, the tracking of the points belonging to the target pattern, found by image matching, was performed with the machine learning model we developed for 10 sequential video images. The features extracted for the machine learning model are: (i) the change between the points of the previous image and the image before that, (ii) the points of interest in the previous image, (iii) the changes found with the homography matrix between sequential images. It was experimentally shown that, point tracking can be achieved with the least error, on avarage about 23 pixels for a 2 mega-pixel resolution image, among the algorithms in the literature that can process more than 30 images per second in a CPU environment of 2 GHz or above.
This paper presents an innovative tool for the study of flame instability phenomena based on high-speed imaging and imageprocessing. The methodology underlying the system exploits the chemiluminescence radiation prod...
详细信息
ISBN:
(纸本)9780791887967
This paper presents an innovative tool for the study of flame instability phenomena based on high-speed imaging and imageprocessing. The methodology underlying the system exploits the chemiluminescence radiation produced by the combustion reactions that occur in flames and which are detected by image sensors of high-speed cameras, speeds compatible with the oscillation frequencies of the phenomena we want to investigate. In general, for gas turbine burners the frequency range of interest varies from a few tens of Hz to a few hundred Hz. According to the author, the most innovative aspect of this methodology is that of providing the results of the analyzes in realtime and not as a post-process activity. To achieve this result, most measurement analysis calculations are performed not on the CPU but on the GPU, drastically reducing calculation times and allowing the presentation of results in realtime. Furthermore, the results presentation part provides a clear visualization of the oscillation modes of light intensity produced by the flame.
Counterfeit medicines present a severe public health threat, especially in low-resource countries where consumers lack reliable means to verify the medicines they purchase. Visual inspection of medicine packaging imag...
详细信息
ISBN:
(纸本)9798331529543;9798331529550
Counterfeit medicines present a severe public health threat, especially in low-resource countries where consumers lack reliable means to verify the medicines they purchase. Visual inspection of medicine packaging images through keypoint matching techniques offers a promising approach for detecting design inconsistencies that could indicate counterfeit products. However, conventional methods often struggle with high computational costs and reduced accuracy when processingimages of varying quality and perspectives. To address these limitations, we propose the Angle and Scale Voting (ASVote) method, which enhances keypoint-based image matching by introducing a 2D voting mechanism that leverages relative angles and scales of the keypoints to eliminate false matches(outliers) while identifying consistent matches (inliers). This approach significantly improves both processingtime and accuracy. Experiments on a real-world dataset of medicine packages show that ASVote improves processingtime and accuracy, outperforming conventional methods.
We present a self-supervised pre-training scheme for single image denoising based on a novel pretext task. Our work is inspired by the success of self-supervised learning (SSL) methods in transfer learning. These meth...
详细信息
ISBN:
(纸本)9798350349405;9798350349399
We present a self-supervised pre-training scheme for single image denoising based on a novel pretext task. Our work is inspired by the success of self-supervised learning (SSL) methods in transfer learning. These methods have been shown to be extremely effective when used to pre-train a model that is then fine-tuned on small datasets. As pretext task, we propose to train a denoising network on patches of the downsampled input image, which we treat as pseudo-clean image patches, and an adaptive noise estimator to learn the specific noise distribution of the input image. By carrying out the pre-training on the single input image, rather than on a separate dataset, we avoid the well-known noise distribution gap between images in the training dataset and the single input image used at test time. We evaluate our SSL method for single image denoising via extensive experiments on both synthetic and real-world noisy image datasets. We demonstrate SotA results compared to existing unsupervised denoising methods, by transferring our pre-training to IDR [1], thus showing that SSL pre-training is a promising framework also in image denoising. Website: https://***/SSL-Denoising.
Targeting UAVs TOPS SAR real-time imaging system, the overall system architecture, imaging algorithm and software architecture design scheme are given. Several engineering implementation problem such as Doppler center...
详细信息
There has been a rise in the frequency of fire-related calamities all over the globe, which leads to the need for an efficient fire detection system to avoid high losses or fatalities. This paper focuses on real-time ...
详细信息
This study introduces a proposed integrated imageprocessing pipeline to enhance vehicle detection and counting precision in real-time video streams. This method can accurately pinpoint areas in the videos where cars ...
详细信息
This study introduces a proposed integrated imageprocessing pipeline to enhance vehicle detection and counting precision in real-time video streams. This method can accurately pinpoint areas in the videos where cars are present by using Regions of Interest (ROIs) and segmenting the frames. To ensure efficient processing of images, it is essential to optimize their quality and dimensions. Utilizing Convolutional Neural Networks (CNNs) for feature extraction enables us to impart discriminative features through hierarchical layers. Afterward, machine learning models enhance the extracted features before applying them to classify automobiles. There are two post-processing tasks: implementing vehicle counting methods that consider discovered ROIs and optimizing image size further. The primary aim is to achieve precise and effective vehicle counting and identification with this method, which is essential for tasks like traffic surveillance. Based on the experimental findings, the system effectively balances processing efficiency and accuracy in vehicle recognition and classification. With this integrated infrastructure, real-time enhancements could be made to processing traffic surveillance video streams.
This study introduces a machine vision system integrated into cyber-physical systems (CPS) for enhanced industrial control. The system employs a specialized script for real-timeimageprocessing and edge detection, wi...
详细信息
ISBN:
(纸本)9798350372977;9798350372984
This study introduces a machine vision system integrated into cyber-physical systems (CPS) for enhanced industrial control. The system employs a specialized script for real-timeimageprocessing and edge detection, with a focus on precision and speed. Results showcase the system's rapid processing capabilities and high-accuracy feature detection, facilitated by machine learning algorithms that enable adaptability and iterative improvement. The system distinguishes itself by not only providing rapid and accurate feature recognition but also by outputting precise coordinates, crucial for micron-level manufacturing precision. An intuitive human-machine interface ensures seamless operation within industrial workflows. This integration significantly improves automated quality control and operational efficiency, demonstrating the system's potential to advance smart manufacturing in line with Industry 4.0 standards.
As the number of imageprocessing increases and the complexity of graphics processing algorithms becomes higher, the real-timeimageprocessing becomes very important. In order to solve the problem of real-time data p...
详细信息
暂无评论