- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Recurrent conditional generative adversarial network for image deblurring
摘要： Nowadays, there is an increasing demand for images with high definition and fine textures, but images captured in natural scenes usually suffer from complicated blurry artifacts, caused mostly by object motion or camera shaking. Since these annoying artifacts greatly decrease image visual quality, deblurring algorithms have been proposed from various aspects. However, most energy-optimization-based algorithms rely heavily on blur kernel priors, and some learning-based methods either adopt pixel-wise loss function or ignore global structural information. Therefore, we propose an image deblurring algorithm based on recurrent conditional generative adversarial network (RCGAN), in which the scale-recurrent generator extracts sequence spatio-temporal features and reconstructs sharp images in a coarse-to-fine scheme. To thoroughly evaluate the global and local generator performance, we further propose a receptive field recurrent discriminator. Besides, the discriminator takes blurry images as conditions, which help to differentiate reconstructed images from real sharp ones. Last but not least, since the gradients are vanishing when training generator with the output of discriminator, a progressive loss function is proposed to enhance the gradients in back-propagation and to take full advantages of discriminative features. Extensive experiments prove the superiority of RCGAN over state-of-the-art algorithms both qualitatively and quantitatively.
关键词： coarse-to-fine,Image deblurring,receptive field recurrent,conditional generative adversarial network
更新于2025-09-23 15:23:52
-
Light Field Image Compression Using Generative Adversarial Network Based View Synthesis
摘要： Light ?eld (LF) has become an attractive representation of immersive multimedia content for simultaneously capturing both the spatial and angular information of the light rays. In this paper, we present a LF image compression framework driven by generative adversarial network (GAN) based sub-aperture image (SAI) generation and cascaded hierarchical coding structure. Speci?cally, we sparsely sample the SAIs in LF and propose the generative adversarial network of LF (LF-GAN) to generate the unsampled SAIs by analogy with adversarial learning conditioned on its surrounding contexts. In particular, LF-GAN learns to interpret both the angular and spatial context of the LF structure, and meanwhile generates intermediate hypothesis for the unsampled SAIs in a certain position. Subsequently, the sampled SAIs and the residues of the generated-unsampled SAIs are re-organized as pseudo-sequences and compressed by standard video codecs. Finally, the hierarchical coding structure is adopted for the sampled SAI to effectively remove the inter-view redundancies. During the training process of LF-GAN, the pixel-wise Euclidean loss as well as the adversarial loss are chosen as the optimization objective, such that sharp textures with less blurring in details can be produced. Extensive experimental results show that the proposed LF-GAN based LF image compression framework outperforms the state-of-the-art learning based LF image compression approach with on average 4.9% BD-rate reductions over multiple LF datasets.
关键词： adversarial learning,SAI synthesis,hierarchical coding,Light ?eld image compression
更新于2025-09-23 15:23:52
-
A Generative Discriminatory Classified Network for Change Detection in Multispectral Imagery
摘要： Multispectral image change detection based on deep learning generally needs a large amount of training data. However, it is difficult and expensive to mark a large amount of labeled data. To deal with this problem, we propose a generative discriminatory classified network (GDCN) for multispectral image change detection, in which labeled data, unlabeled data, and new fake data generated by generative adversarial networks are used. The GDCN consists of a discriminatory classified network (DCN) and a generator. The DCN divides the input data into changed class, unchanged class, and extra class, i.e., fake class. The generator recovers the real data from input noises to provide additional training samples so as to boost the performance of the DCN. Finally, the bitemporal multispectral images are input to the DCN to get the final change map. Experimental results on the real multispectral imagery datasets demonstrate that the proposed GDCN trained by unlabeled data and a small amount of labeled data can achieve competitive performance compared with existing methods.
关键词： Change detection,deep learning,multispectral imagery,generative adversarial networks (GANs)
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
-
[IEEE 2018 25th International Conference on Mechatronics and Machine Vision in Practice (M2VIP) - Stuttgart, Germany (2018.11.20-2018.11.22)] 2018 25th International Conference on Mechatronics and Machine Vision in Practice (M2VIP) - Generative models for direct generation of CNC toolpaths
摘要： Today, numerical controls (CNC) are the standard for the control of machine tools and industrial robots in production and enable highly flexible and efficient production, especially for frequently changing production tasks. A numerical control has discrete inputs and outputs. Within the NC channel, however, it is necessary to analytically describe curves for the calculation of the position setpoints and the jerk limitation. The resulting change between discrete and continuous description forms and the considerable restrictions in the parallelisation of the interpolation of continuous curves within the NC channel lead to a performance overhead that limits the performance of the NC channel with regard to the calculation of new position setpoints. This can lead to a drop in production speed and thus to longer production times. To solve this problem, we propose a new approach in this paper. This is based on the use of deep generative models and allows the direct generation of interpolated toolpaths without calculation of continuous curves and subsequent discretization. The generative models are being trained to create curves of certain types such as linear and parabolic curves or splines directly as discrete point sequences. This approach is very well feasible with regard to its parallelization and reduces the computing effort within the NC channel. First results with straight lines and parabolic curves show the feasibility of this new approach for the generation of CNC toolpaths.
关键词： machine learning,computerized numerical control,interpolation,CNC,generative adversarial networks
更新于2025-09-23 15:22:29
-
Task-Oriented GAN for PolSAR Image Classification and Clustering
摘要： Based on a generative adversarial network (GAN), a novel version named Task-Oriented GAN is proposed to tackle difficulties in PolSAR image interpretation, including PolSAR data analysis and small sample problem. Besides two typical parts in GAN, i.e., generator (G-Net) and discriminator (D-Net), there is a third part named TaskNet (T-Net) in the Task-Oriented GAN, where T-Net is employed to accomplish a certain task. Two tasks, PolSAR image classification and clustering, are studied in this paper, where T-Net acts as a Classifier and a Clusterer, respectively. The learning procedure of Task-Oriented GAN consists of two main stages. In the first stage, G-Net and D-Net vie with each other like that in a general GAN; in the second stage, G-Net is adjusted and oriented by T-Net so that more samples, which are benefit for the task and called fake data, are generated. As a result, Task-Oriented GAN not only has the advantage of GAN (no-assumption data modeling) but also overcomes the disadvantage of GAN (task-free). After learning, fake data are employed to enrich training set and avoid overfitting; so Task-Oriented GAN performs well even if the manual-labeled data are small. To verify the effectiveness of T-Net, a visualized comparison is provided, where some fake digits generated from Task-Oriented GAN are illustrated along with that from GAN. What is more, considering that there is a great difference between PolSAR data and general data, in our PolSAR image classification and clustering tasks, the specific PolSAR information is inserted into the structure of the Task-Oriented GAN. This enables researchers to mine inherent information in PolSAR data without any data hypothesis and find ways for small sample problem at the same time. Experiment results tested on three PolSAR images show that the proposed method performs well in dealing with PolSAR image classification and clustering.
关键词： generative adversarial network (GAN),task-oriented,Clustering,PolSAR image classification
更新于2025-09-23 15:22:29
-
Faulty elements diagnosis of phased array antennas using a generative adversarial learning-based stacked denoising sparse autoencoder
摘要： Diagnosis of faulty elements in a linear phased array antenna is of great importance in the wireless communication field which has been received increasing attention. As a result of element or elements failure in the linear phased array antennas, the whole radiation pattern will suffer from high side lobe levels, wide bandwidth and unexpected nulls. To this end, we suggest a novel approach by combining the generative adversarial learning and the stacked denoising sparse autoencoder to determine the location of the faulty elements in antennas. The suggested approach can learn discriminative features from radiation pattern images adaptively and automatically with less expert knowledge. Meanwhile, the suggested approach is able to overcome the strong noise, the high dimensional size of the radiation pattern and the small fault samples. In this regard, the suggested approach possesses superiority in discriminant capability in contrast to the existing related approaches.
关键词： stacked denoising sparse autoencoder,phased array antennas,Faulty elements diagnosis,generative adversarial learning
更新于2025-09-23 15:22:29
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - Unsupervised Multi-Domain Image Translation with Domain-Specific Encoders/Decoders
摘要： Unsupervised Image-to-Image Translation achieves spectacularly advanced developments nowadays. However, recent approaches mainly focus on one model with two domains, which may face heavy burdens with the large cost of training time and the huge model parameters, under such a requirement that n (n>2) domains are freely transferred to each other in a general setting. To address this problem, we propose a novel and unified framework named Domain-Bank, which consists of a globally shared auto-encoder and n domain-specific encoders/decoders, assuming that there is a universal shared-latent space can be projected. Thus, we not only reduce the parameters of the model but also have a huge reduction of the time budgets. Besides the high efficiency, we show the comparable (or even better) image translation results over state-of-the-arts on various challenging unsupervised image translation tasks, including face image translation and painting style translation. We also apply the proposed framework to the domain adaptation task and achieve state-of-the-art performance on digit benchmark datasets.
关键词： Shared-Latent Space,Unsupervised Image-to-Image Translation,Generative Adversarial Networks,Variational Autoencoders,Domain-Bank,Multi-Domain
更新于2025-09-23 15:22:29
-
[IEEE 2018 5th NAFOSTED Conference on Information and Computer Science (NICS) - Ho Chi Minh, Vietnam (2018.11.23-2018.11.24)] 2018 5th NAFOSTED Conference on Information and Computer Science (NICS) - Joint Image Deblurring and Binarization for License Plate Images using Deep Generative Adversarial Networks
摘要： Image deblurring is a highly ill-posed inverse problem where it aims to estimate the sharp image from blurred image with or without the knowledge about the blurring process. Despite the success of model-based image deblurring methods where the deconvolution is a major step to recover the sharp image, its usage in practice is still limited, especially when many factors such as object motion, camera motion, non-uniform sensitivity of the imaging device contribute to imaging process. In automatic license plate recognition (ALPR) of moving vehicle, the blurred image severely reduces the accuracy of recognition. Meanwhile, though the binarized image of license plate has an important role in ALPR systems, its accuracy is largely affected by the blurred image. In this paper, we use a deep architecture based on Generative Adversarial Networks to jointly perform image deblurring and image binarization for license plate images. Our model directly maps from blurred image to binary image without going through the deblurring as in conventional method. The proposed method is benefited from the fact that the ground-truth, sharp license plates are difficult to acquire for moving object, while the accurate binary images can be manually derived from blurred ones.
关键词： Inverse Problems,License Plate Deblurring,Image Deblurring,Generative Adversarial Network (GAN)
更新于2025-09-23 15:22:29
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Road Segmentation of UAV RS Image Using Adversarial Network with Multi-Scale Context Aggregation
摘要： Semantic segmentation using adversarial networks has been approved to produce the better artificial results in image processing fields. Focused on current Deep Convolutional Neural Networks (DCNNs), since the convolutional kernel size has been fixed in every convolutional operation, the small objects would be ignored with large convolutional kernel size, and the segmentation result of large objects is not continuous with small convolutional kernel size. The paper developed a semantic segmentation model that combined the adversarial networks with multi-scale context aggregation. Further, the model was applied to road segmentation of UAV RS images. The experimental results of this semantic segmentation model with multi-scale context aggregation has a better performance for road segmentation and fit well with the reference standard results. It can improve the road segmentation accuracy obviously in the situation where there are other small regions whose shape or color is similar to road regions in UAV RS images.
关键词： Road Segmentation,Adversarial Network,UAV image,Image processing,multi-scale context aggregation
更新于2025-09-23 15:22:29