- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Non-blind image deblurring method by the total variation deep network
摘要： There are a lot of non-blind image deblurring methods, especially, with the total variation (TV) model-based method. However, how to choose the parameters adaptively for regularization is a major open problem. We proposed a very novel method which is based on TV deep network to learn the best parameters adaptively for regularization. We used deep learning and prior knowledge to set up a TV-based deep network and calculate parameters of regularization such as biases and weights. Therefore, we used the idea of a deep network to update these parameters automatically so that avoiding sophisticated calculations. Our experimental results by our proposed network are significantly better than several other methods, in respect of detail retention and anti-noise performance. At the same time, we can achieve the same effect with a minimum number of training sets, thus speeding up the calculation.
关键词： Total variation model,deep learning,Non-blind image deblurring
更新于2025-09-23 15:23:52
-
[IEEE 2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE) - Vilnius, Lithuania (2018.11.8-2018.11.10)] 2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE) - Automated Image Annotation based on YOLOv3
摘要： A typical pedestrian protection system requires sophisticated hardware and robust detection algorithms. To solve these problems the existing systems use hybrid sensors where mono and stereo vision merged with active sensors. One of the most assuring pedestrian detection sensors is far infrared range camera. The classical pedestrian detection approach based on Histogram of oriented gradients is not robust enough to be applied in devices which consumers can trust. An application of deep neural network-based approach is able to perform with significantly higher accuracy. However, the deep learning approach requires a high number of labeled data examples. The investigation presented in this paper aimed the acceleration of pedestrian labeling in far-infrared image sequences. In order to accelerate pedestrian labeling in far-infrared camera videos, we have integrated the YOLOv3 object detector into labeling software. The verification of the pre-labeled results was around eleven times faster than manual labeling of every single frame.
关键词： deep-learning,YOLOv3,Far-infrared,pedestrian detection,annotation labeling
更新于2025-09-23 15:23:52
-
Single infrared image enhancement using a deep convolutional neural network
摘要： In this paper, we propose a deep learning method for single infrared image enhancement. A fully convolutional neural network (CNN) is used to produce images with enhanced contrast and details. The conditional generative adversarial networks are incorporated into the optimization framework to avoid the background noise being amplified and further enhance the contrast and details. The existing convolutional neural network architectures, such as residual architectures and encoder–decoder architectures, fail to achieve the best results both in terms of network performance and application scope for infrared image enhancement task. To address this problem, we specifically design a new refined convolutional neural architecture that produces visually very appealing results with higher contrast and sharper details compared to other network architectures. Visible images are used for training since there are fewer infrared images. Proper training samples are generated to ensure that the network trained on visible images can be well applied to infrared images. Experiments demonstrate that our approach outperforms existing image enhancement algorithms in terms of contrast and detail enhancement. Code is available at https://github.com/Kuangxd/IE-CGAN.
关键词： Residual network,Enhancement,Infrared images,Deep learning,Encoder–decoder network,Generative adversarial network
更新于2025-09-23 15:23:52
-
A review of image-based automatic facial landmark identification techniques
摘要： The accurate identification of landmarks within facial images is an important step in the completion of a number of higher-order computer vision tasks such as facial recognition and facial expression analysis. While being an intuitive and simple task for human vision, it has taken decades of research, an increase in the availability of quality data sets, and a dramatic improvement in computational processing power to achieve near-human accuracy in landmark localisation. The intent of this paper is to provide a review of the current facial landmarking literature, outlining the significant progress that has been made in the field from classical generative methods to more modern techniques such as sophisticated deep neural network architectures. This review considers a generalised facial landmarking problem and provides experimental examples for each stage in the process, reporting repeatable benchmarks across a number of publicly available datasets and linking the results of these examples to the recently reported performance in the literature.
关键词： Vision,Landmarking,Face,Registration,Survey,Image,Review,Artificial neural networks,Deep learning,Machine learning
更新于2025-09-23 15:22:29
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - Learning Illuminant Estimation from Object Recognition
摘要： In this paper we present a deep learning method to estimate the illuminant of an image. Our model is not trained with illuminant annotations, but with the objective of improving performance on an auxiliary task such as object recognition. To the best of our knowledge, this is the first example of a deep learning architecture for illuminant estimation that is trained without ground truth illuminants. We evaluate our solution on standard datasets for color constancy, and compare it with state of the art methods. Our proposal is shown to outperform most deep learning methods in a cross-dataset evaluation setup, and to present competitive results in a comparison with parametric solutions.
关键词： Illuminant estimation,deep learning,convolutional neural networks,computational color constancy,semi-supervised learning
更新于2025-09-23 15:22:29
-
Context-Aware Depth and Pose Estimation for Bronchoscopic Navigation
摘要： Endobronchial intervention is increasingly used as a minimally invasive means of lung intervention. Vision-based localization approaches are often sensitive to image artifacts in bronchoscopic videos. In this paper, a robust navigation system based on a context-aware depth recovery approach for monocular video images is presented. To handle the artifacts, a conditional generative adversarial learning framework is proposed for reliable depth recovery. The accuracy of depth estimation and camera localization is validated on an in vivo dataset. Both quantitative and qualitative results demonstrate that the depth recovered with the proposed method preserves better structural information of airway lumens in the presence of image artifacts, and the improved camera localization accuracy demonstrates its clinical potential for bronchoscopic navigation.
关键词： Computer Vision for Medical Robotics,Deep Learning in Robotics and Automation,Visual-Based Navigation,Visual Learning
更新于2025-09-23 15:22:29
-
[IEEE 2018 3rd International Conference on Mechanical, Control and Computer Engineering (ICMCCE) - Huhhot (2018.9.14-2018.9.16)] 2018 3rd International Conference on Mechanical, Control and Computer Engineering (ICMCCE) - A Remote Sensing Image Key Target Recognition System Design Based on Faster R-CNN
摘要： Aiming at the problem of traditional low-level recognition of key targets in remote sensing images, a method for target detection and recognition based on Faster R-CNN is proposed. Firstly, the open source remote sensing image data set NWPU VHR-10 dataset is converted into VOC 2007 format as the training sets and test sets. Secondly, according to the training set category information, the hyper-parameters of the neural network are refined, and then the training set is trained using the Faster R-CNN neural network to generate a model. Finally, this model is used to detect unknown remote sensing images and identify important targets. The simulation results show that the method has high recognition accuracy and speed, and can provide reference for recognition of the key targets of remote sensing images.
关键词： Faster R-CNN,convolution neural network,deep learning,key target recognition,remote sensing image detection
更新于2025-09-23 15:22:29
-
PolSAR Image Semantic Segmentation Based on Deep Transfer Learning--Realizing Smooth Classification With Small Training Sets
摘要： Suffering from speckle noise and complex scattering phenomena, classification results of SAR images are usually noisy and shattered, which makes them difficult to use in practical applications. Deep-learning-based semantic segmentation realizes segmentation and categorization at the same time, and thus can obtain smooth and fine-grained classification maps. However, this kind of methods require large data sets with pixel-wise categorical annotations, which are time consuming and tedious to retrieve. Compared with photographs and optical remote sensing images, manually annotating SAR data is even harder, which results in a delay of using relevant techniques in this field. In this letter, a new data set is proposed to support semantic segmentation for high-resolution PolSAR images. Limited by the aforementioned problems, the data set is only a small one with 50 image patches. Therefore, two transfer learning strategies are proposed, which adopt the fully convolutional network (FCN) and U-net architecture, respectively, and use distinct pretraining data sets to adapt to different situations. The experiments demonstrate the good performance of both methods and a promising applicability of using small training sets. Moreover, although trained with small patches, both networks can perfectly apply on large images. The new data set and methods are hopeful to support various PolSAR applications as baselines.
关键词： polarimetry,SAR,image classification,Deep learning,image segmentation
更新于2025-09-23 15:22:29
-
A Novel Patch Variance Biased Convolution Neural Network for No-Reference Image Quality Assessment
摘要： Deep Convolutional Neural Networks (CNNs) have been successfully applied on no-reference image quality assessment (NR-IQA) with respect to human perception. Most of these methods deal with small image patches and use the average score of the test patches for predicting the whole image quality. We discovered that image patches from homogenous regions are unreliable for both neural network training and final image quality score estimation. In addition, image patches with complex structures have much higher chances to achieve better image quality prediction. Based on these findings, we enhanced the conventional CNN-based NR-IQA algorithm to avoid homogenous patches for the network training and quality score estimation. Moreover, we also use a variance-based weighting average to bias the final image quality score to the patches with complex structure. Experimental results show that this simple approach can achieve state-of-the-art performance as compared with well-known NR-IQA algorithms.
关键词： deep learning,no-reference image quality assessment,convolution neural network
更新于2025-09-23 15:22:29
-
Semisupervised Stacked Autoencoder With Cotraining for Hyperspectral Image Classification
摘要： Recently, deep learning (DL) is of great interest in hyperspectral image (HSI) classification. Although many effective frameworks exist in the literature, the generally limited availability of training samples poses great challenges in applying DL to HSI classification. In this paper, we present a novel DL framework, namely, semisupervised stacked autoencoders (Semi-SAEs) with cotraining, for HSI classification. First, two SAEs are pretrained based on the hyperspectral features and the spatial features, respectively. Second, fine-tuning is alternatively conducted for the two SAEs in a semisupervised cotraining fashion, where the initial training set is enlarged by designing an effective region growing method. Finally, the classification probabilities obtained by the two SAEs are fused using a Markov random field model solved by iterated conditional modes. Experimental results based on three popular hyperspectral data sets demonstrate that the proposed method outperforms other state-of-the-art DL methods.
关键词： Deep learning (DL),stacked autoencoders (SAEs),cotraining,hyperspectral image (HSI) semisupervised classification,Markov random field (MRF)
更新于2025-09-23 15:22:29