Quantum imageprocessing is paramount in imageprocessing-based research areas nowadays due to the less computational time. This paper presents the quantum mechanic-based image denoising model. Also, This article disc...
详细信息
The proceedings contain 27 papers. The topics discussed include: improved adaptive threshold segmentation of ultrasound medical images;a real-time video analysis software based on deep convolutional neural networks (D...
ISBN:
(纸本)9781450390057
The proceedings contain 27 papers. The topics discussed include: improved adaptive threshold segmentation of ultrasound medical images;a real-time video analysis software based on deep convolutional neural networks (DCNN) for useful and effective endoscopy video storage;three-dimensional medical image segmentation with SE-VNet neural networks;spine curve extraction based on mask segmentation;on the use of boundary gradient for the analysis of MR wrist bones volumes segmentation;depression recognition method based on regional homogeneity features from emotional response fMRI using deep convolutional neural network;automatic discrimination of fundus DR based on improved residual dense block network;multi-tissue derived windowing technology based on statistical features and its colorization application;preserving the temporal consistency of video sequences for surgical instruments segmentation;and identifying biomarkers of HCV-induced dysplasia and hepatocellular carcinoma based on network centrality.
With the rapid development of digital media technology, image analysis plays a vital role in the fields of digitalization and informatization. This article will provide an in-depth explanation of face recognition and ...
详细信息
Drowsiness among drivers poses an urgent safety problem in road transportation because it creates serious risks to both drivers and their passengers. Carrying out prolonged driving operations creates a mind state betw...
详细信息
Adversarial examples, deliberately crafted using small perturbations to fool deep neural networks, were first studied in imageprocessing and more recently in NLP. While approaches to detecting adversarial examples in...
详细信息
ISBN:
(纸本)9798891760189
Adversarial examples, deliberately crafted using small perturbations to fool deep neural networks, were first studied in imageprocessing and more recently in NLP. While approaches to detecting adversarial examples in NLP have largely relied on search over input perturbations, imageprocessing has seen a range of techniques that aim to characterise adversarial subspaces over the learned representations. In this paper, we adapt two such approaches to NLP, one based on nearest neighbors and influence functions and one on Mahalanobis distances. The former in particular produces a state-of-the-art detector when compared against several strong baselines;moreover, the novel use of influence functions provides insight into how the nature of adversarial example subspaces in NLP relate to those in imageprocessing, and also how they differ depending on the kind of NLP task.
In terms of ensuring the safety of construction workers and cyclists, traditional positioning methods often rely on GPS signals, but the accuracy of GPS positioning is limited indoors or in signal-blocked areas. At th...
详细信息
Accurate classification of land cover from aerial images is one of the research topics in remote sensing and is also in high demand in industry. However, obtaining labeled data for training different classifiers that ...
详细信息
ISBN:
(纸本)9798350350494;9798350350500
Accurate classification of land cover from aerial images is one of the research topics in remote sensing and is also in high demand in industry. However, obtaining labeled data for training different classifiers that heavily depend on supervision is still a challenging and resource-intensive task. Unsupervised methods have emerged as a powerful alternative to overcome the limitations associated with labeled data. Such methods have a high ability to discover hidden patterns and structures in multi-spectral images and have the possibility of classifying various types of land cover without relying on labeled samples. Our research primarily involved the analysis of World-View3 satellite imagery. Our strategy involved creating an advanced pipeline that extracted features using autoencoders. Through this approach, the multi-spectral images' key characteristics are efficiently extracted. Subsequently, we implement transfer learning to re-train the model with a limited number of labeled data. By applying transfer learning, our pipeline significantly enhances the capability of multispectral imageprocessing, enabling a more comprehensive and accurate interpretation of satellite imagery data. Finally, we evaluate our results not only by providing a confusion matrix but also through a visual comparison between the class map and the RGB composition of the MSI image.
Smart farming has become essential, and the use of Unmanned Aerial Vehicle (UAV) photography is becoming an essential component of the process. To successfully acquire UAV footage of different spectral bands and ident...
详细信息
This study aims to explore the grotto color analysis based on imageprocessing and its application in environmental design. By using multispectral and hyperspectral imaging techniques, we collected and analyzed the co...
详细信息
This research paper explores the development and implementation of 'NetraAI - The 3rd Eye,' an AI-powered surveillance system aimed at enhancing public safety and security measures. The study investigates the ...
详细信息
暂无评论