scholarly journals Multi-Scale Fused SAR Image Registration Based on Deep Forest

2021 ◽  
Vol 13 (11) ◽  
pp. 2227
Author(s):  
Shasha Mao ◽  
Jinyuan Yang ◽  
Shuiping Gou ◽  
Licheng Jiao ◽  
Tao Xiong ◽  
...  
Keyword(s):  
Image Registration ◽  
Multiple Scales ◽  
Binary Classification ◽  
Reference Image ◽  
Sar Image ◽  
Multi Scale ◽  
Training Samples ◽  
Key Points ◽  
Deep Forest ◽  
Matching Models

SAR image registration is a crucial problem in SAR image processing since the registration results with high precision are conducive to improving the quality of other problems, such as change detection of SAR images. Recently, for most DL-based SAR image registration methods, the problem of SAR image registration has been regarded as a binary classification problem with matching and non-matching categories to construct the training model, where a fixed scale is generally set to capture pair image blocks corresponding to key points to generate the training set, whereas it is known that image blocks with different scales contain different information, which affects the performance of registration. Moreover, the number of key points is not enough to generate a mass of class-balance training samples. Hence, we proposed a new method of SAR image registration that meanwhile utilizes the information of multiple scales to construct the matching models. Specifically, considering that the number of training samples is small, deep forest was employed to train multiple matching models. Moreover, a multi-scale fusion strategy is proposed to integrate the multiple predictions and obtain the best pair matching points between the reference image and the sensed image. Finally, experimental results on four datasets illustrate that the proposed method is better than the compared state-of-the-art methods, and the analyses for different scales also indicate that the fusion of multiple scales is more effective and more robust for SAR image registration than one single fixed scale.

SAR Image Registration Using Ratio Mutual Information

2012 ◽  
Vol 241-244 ◽  
pp. 2630-2637
Author(s):  
Chun Rong Wei ◽  
Chu He ◽  
Hong Sun
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
Speckle Noise ◽  
Synthetic Aperture Radar Image ◽  
Reference Image ◽  
Sar Image ◽  
Local Maxima ◽  
Characteristic Quantity ◽  
Registration Algorithm ◽  
Image Registration Algorithm

In order to reduce the noise sensitivity of the SAR (synthetic aperture radar) image registration, a image registration algorithm which basing on the ratio mutual information (RatioMI) is proposed in this paper. Firstly, the ratio images of the reference image and the floating image are gotten by using the ratio operator, and then take the two ratio images as a similar characteristic quantity to construct the similarity measure function which was used in the optimization process of the image registration experiment. The experimental results of the SAR image registration show that the new registration algorithm which based on the RatioMI is effectively in avoiding the local maxima point problems causing by speckle noise.

A Modified Phase Correlation Method for Image Registration

2011 ◽  
Vol 48-49 ◽  
pp. 48-51
Author(s):  
Lu Jing Yang ◽  
Wei Hao ◽  
Chong Lun Li
Keyword(s):  
Image Registration ◽  
Rotation Angle ◽  
Translational Motion ◽  
Correlation Method ◽  
Phase Correlation ◽  
Flow Chart ◽  
Reference Image ◽  
Sar Image ◽  
Registration Method ◽  
Simulation Results

Image registration is a very fundamental and important part in many multi-sensor image based applications. Phase correlation-based image registration method is widely concerned for its small computation amount, strong anti-interference property. However, it can only solve the image registration problem with translational motion. Hence, we proposed a modified phase correlation registration method in the paper. We analyzed the principle of registration, gave the flow chart, and applied the method to the SAR image registration problems with scaling, rotation and translation transformation. Simulation results show that the method can accurately estimate the translation parameters, zoom scale and rotation angle of registrating image relative to the reference image.

Development of New Image Registration Techniques – A Research Framework

2018 ◽  
Vol 8 (3) ◽  
pp. 50
Author(s):  
Sindhu Madhuri G. ◽  
Indira Gandhi M P
Keyword(s):  
Image Processing ◽  
Image Registration ◽  
Digital Image Processing ◽  
Digital Image ◽  
Creative Thinking ◽  
Reference Image ◽  
Transformation Technique ◽  
Research Framework ◽  
Registration Process ◽  
Data Source

Image is a basic and fundamental data source for the digital image processing. This image data source is required to be processed into information or intelligence and further to knowledge levels where it is required to understand and migrate into knowledge economy systems. Image registration is one of such key and most important process already identified in the digital image processing domain. Image registration is a process of bringing the reference image and sensed image into a common co-ordinate system, and application of complex transformation techniques for necessary comparison of reference with sensed images obtained from different - views, times, spaces, etc., in order to extract the valuable information and intelligence embedded in them. Due to the complexity of overall image registration process, it is difficult to suggest a single transformation technique even for a specific application. In addition, it is highly impossible to suggest one single transformation technique for comparison of various sensed images with a reference image during the image registration process. This research gap calls for the development of new image registration techniques for the application of more than one transformation technique during the image registration process for the necessary comparisons with reference image & sensed images, those are obtained from the available heterogeneous sources or sensors, based on the requirement. In addition, it is a basic need to attempt for the measurement of effectiveness of the image registration process also. Therefore, a research framework is developed for image registration process and attempted for the measurement of its effectiveness also. This new research area is a novel idea, and is expected to emerge as a provision for the knowledge computations with creative thinking through the embedded intelligence extraction during the complex image registration process.

scholarly journals Target Detection Network for SAR Images Based on Semi-Supervised Learning and Attention Mechanism

2021 ◽  
Vol 13 (14) ◽  
pp. 2686
Author(s):  
Di Wei ◽  
Yuang Du ◽  
Lan Du ◽  
Lu Li
Keyword(s):  
Feature Extraction ◽  
Supervised Learning ◽  
Target Detection ◽  
Scene Recognition ◽  
Attention Mechanism ◽  
Sar Image ◽  
Target Level ◽  
Sar Images ◽  
Training Samples ◽  
Pass Through

The existing Synthetic Aperture Radar (SAR) image target detection methods based on convolutional neural networks (CNNs) have achieved remarkable performance, but these methods require a large number of target-level labeled training samples to train the network. Moreover, some clutter is very similar to targets in SAR images with complex scenes, making the target detection task very difficult. Therefore, a SAR target detection network based on a semi-supervised learning and attention mechanism is proposed in this paper. Since the image-level label simply marks whether the image contains the target of interest or not, which is easier to be labeled than the target-level label, the proposed method uses a small number of target-level labeled training samples and a large number of image-level labeled training samples to train the network with a semi-supervised learning algorithm. The proposed network consists of a detection branch and a scene recognition branch with a feature extraction module and an attention module shared between these two branches. The feature extraction module can extract the deep features of the input SAR images, and the attention module can guide the network to focus on the target of interest while suppressing the clutter. During the semi-supervised learning process, the target-level labeled training samples will pass through the detection branch, while the image-level labeled training samples will pass through the scene recognition branch. During the test process, considering the help of global scene information in SAR images for detection, a novel coarse-to-fine detection procedure is proposed. After the coarse scene recognition determining whether the input SAR image contains the target of interest or not, the fine target detection is performed on the image that may contain the target. The experimental results based on the measured SAR dataset demonstrate that the proposed method can achieve better performance than the existing methods.

scholarly journals No-Reference Image Quality Assessment with Multi-Scale Orderless Pooling of Deep Features

2021 ◽  
Vol 7 (7) ◽  
pp. 112
Author(s):  
Domonkos Varga
Keyword(s):  
Image Quality ◽  
Quality Assessment ◽  
Image Quality Assessment ◽  
Multiple Scales ◽  
Digital Images ◽  
Input Image ◽  
Perceptual Quality ◽  
Reference Image ◽  
Benchmark Datasets ◽  
In The Wild

The goal of no-reference image quality assessment (NR-IQA) is to evaluate their perceptual quality of digital images without using the distortion-free, pristine counterparts. NR-IQA is an important part of multimedia signal processing since digital images can undergo a wide variety of distortions during storage, compression, and transmission. In this paper, we propose a novel architecture that extracts deep features from the input image at multiple scales to improve the effectiveness of feature extraction for NR-IQA using convolutional neural networks. Specifically, the proposed method extracts deep activations for local patches at multiple scales and maps them onto perceptual quality scores with the help of trained Gaussian process regressors. Extensive experiments demonstrate that the introduced algorithm performs favorably against the state-of-the-art methods on three large benchmark datasets with authentic distortions (LIVE In the Wild, KonIQ-10k, and SPAQ).

Sign in / Sign up

Export Citation Format

Share Document