Saliency analysis is essential to detect common regions of interest (ROI) in remote sensing images. However, many methods imply saliency analysis in single images and cannot detect common ROI accurately. In this paper, we propose the joint saliency analysis based on iterative clustering (JSIC) method to detect common ROIs. Firstly, the size of superpixel patch is adaptively determined by texture feature. Secondly, color feature and intensity feature are utilized to get initial saliency maps and Otsu is utilized to obtain initial ROIs. Finally, iterative clustering is applied to obtain final ROI with less background inference. Quantitative and qualitative experiments results show that the iterative clustering joint saliency analysis method not only has better performance when compared to the other state-of-the-art methods, but also can eliminate image without ROI. Our contributions lie in three aspects as follows: 1) We propose a novel method to calculate the number of superpixel blocks adaptively. 2) A new joint saliency analysis method is proposed based on color feature and intensity feature. 3) We propose a novel saliency modification strategy based on the iterative cluster, which could reduce the background inference and eliminate images without ROIs.
Salt-and-pepper (SAP) noise is one of the common impulse noises. It is generated mostly during the process of image capture and storage, due to false locations in memory and damaged image sensors. SAP noise seriously degrades the quality of images and affects the performance of subsequent image processing, such as edge detection, image segmentation and object recognition. Thus, it is quite necessary to remove SAP noise from corrupted images efficiently. In this paper, we propose a fast-adaptive Gaussian-weighted mean filter (FAGWMF) for removing salt-and-pepper noises. Our denoising filter consists of four stages. At the first stage, we preprocess the image by enlarging and flipping the image. Then, we detect noisy pixels by comparing the pixel value with the maximum (255) and minimum (0) value. One pixel is regarded as noise if its value is equal to the maximum or minimum value. Otherwise, it is regarded as noisefree pixel. At the third stage, we determine the working window size by enlarging the filter window continuously until the quantity of noise-free pixels it includes reaches to the predetermined threshold or the window radius reaches to the predetermined maximum value. At the last stage, we replace each noise candidate by its Gaussian-weighted mean value of the noise-free pixels in window. Using Gaussian-weighted template, the central candidates will get larger weights than those on edges, which helps to preserve the edge information efficiently. Simulation results show that compared to some state-of-the-art algorithms, our proposed filter has faster execution speed and better restoration quality.
Target extraction is one of the important aspects in remote sensing image analysis and processing, which has wide applications in images compression, target tracking, target recognition and change detection. Among different targets, airport has attracted more and more attention due to its significance in military and civilian. In this paper, we propose a novel and reliable airport object extraction model combining visual attention mechanism and parallel line detection algorithm. First, a novel saliency analysis model for remote sensing images with airport region is proposed to complete statistical saliency feature analysis. The proposed model can precisely extract the most salient region and preferably suppress the background interference. Then, the prior geometric knowledge is analyzed and airport runways contained two parallel lines with similar length are detected efficiently. Finally, we use the improved Otsu threshold segmentation method to segment and extract the airport regions from the salient map of remote sensing images. The experimental results demonstrate that the proposed model outperforms existing saliency analysis models and shows good performance in the detection of the airport.
Vehicle tracking technology is currently one of the most active research topics in machine vision. It is an important part of intelligent transportation system. However, in theory and technology, it still faces many challenges including real-time and robustness. In video surveillance, the targets need to be detected in real-time and to be calculated accurate position for judging the motives. The contents of video sequence images and the target motion are complex, so the objects can’t be expressed by a unified mathematical model. Object-tracking is defined as locating the interest moving target in each frame of a piece of video. The current tracking technology can achieve reliable results in simple environment over the target with easy identified characteristics. However, in more complex environment, it is easy to lose the target because of the mismatch between the target appearance and its dynamic model. Moreover, the target usually has a complex shape, but the tradition target tracking algorithm usually represents the tracking results by simple geometric such as rectangle or circle, so it cannot provide accurate information for the subsequent upper application. This paper combines a traditional object-tracking technology, Mean-Shift algorithm, with a kind of image segmentation algorithm, Active-Contour model, to get the outlines of objects while the tracking process and automatically handle topology changes. Meanwhile, the outline information is used to aid tracking algorithm to improve it.
The airport is one of the most crucial traffic facilities in military and civil fields. Automatic airport extraction in high spatial resolution remote sensing images has many applications such as regional planning and military reconnaissance. Traditional airport extraction strategies usually base on prior knowledge and locate the airport target by template matching and classification, which will cause high computation complexity and large costs of computing resources for high spatial resolution remote sensing images. In this paper, we propose a novel automatic airport extraction model based on saliency region detection, airport runway extraction and adaptive threshold segmentation. In saliency region detection, we choose frequency-tuned (FT) model for computing airport saliency using low level features of color and luminance that is easy and fast to implement and can provide full-resolution saliency maps. In airport runway extraction, Hough transform is adopted to count the number of parallel line segments. In adaptive threshold segmentation, the Otsu threshold segmentation algorithm is proposed to obtain more accurate airport regions. The experimental results demonstrate that the proposed model outperforms existing saliency analysis models and shows good performance in the extraction of the airport.
The human visual system can quickly focus on a small number of salient objects. This process was known as visual saliency analysis and these salient objects are called focus of attention (FOA). The visual saliency analysis mechanism can be used to extract the salient regions and analyze saliency of object in an image, which is time-saving and can avoid unnecessary costs of computing resources. In this paper, a novel visual saliency analysis model based on dynamic multiple feature combination strategy is introduced. In the proposed model, we first generate multi-scale feature maps of intensity, color and orientation features using Gaussian pyramids and the center-surround difference. Then, we evaluate the contribution of all feature maps to the saliency map according to the area of salient regions and their average intensity, and attach different weights to different features according to their importance. Finally, we choose the largest salient region generated by the region growing method to perform the evaluation. Experimental results show that the proposed model cannot only achieve higher accuracy in saliency map computation compared with other traditional saliency analysis models, but also extract salient regions with arbitrary shapes, which is of great value for the image analysis and understanding.
Accurate region of interest (ROI) extraction is a hotspot of remote sensing image analysis. In this paper, we propose a novel ROI extraction method based on multi-scale hybrid visual saliency analysis (MHVSA) that can be divided into two sub-models: the frequency feature analysis (FFA) model and the multi-scale region aggregation (MRA) model. In the FFA sub-model, we utilize the human visual sensitivity and the Fourier transform to produce the local saliency map. In the MRA sub-model, saliency maps of various scales are generated by aggregating regions. A tree-structure graphical model is suggested to fuse saliency maps into one global saliency map. We obtain two binary masks by segmenting the local and global saliency maps and perform the logical AND operation on the two masks to acquire the final mask. Experimental results reveal that the MHVSA model provides more accurate extraction results.
Region of Interest (ROI) extraction is an important component in remote sensing images processing, which is useful for further practical applications such as image compression, image fusion, image segmentation and image registration. Traditional ROI extraction methods are usually prior knowledge-based and depend on a global searching solution which are time consuming and computational complex. Saliency detection which is widely used for ROI extraction from natural scene images in these years can effectively solve the problem of high computation complexity in ROI extraction for remote sensing images as well as retain accuracy. In this paper, a new computational model is proposed to improve the accuracy of ROI extraction in remote sensing images. Considering the characteristics of remote sensing images, we first use lifting wavelet transform based on adaptive direction evaluation (ADE) to obtain multi-scale orientation contrast feature map (MF). Secondly, the features of color are exploited using the information content analysis to provide a color information map (CIM). Thirdly, feature fusion is used to integrate multi-scale orientation contrast features and color information for generating a saliency map. Finally, an adaptive threshold segmentation algorithm is employed to obtain the ROI. Compared with existing models, our method can not only effectively extract detail of the ROIs, but also effectively remove mistaken detection of the inner parts of the ROIs.
Region of interest (ROI) extraction is an important component of remote sensing image processing. However, traditional ROI extraction methods are usually prior knowledge-based and depend on classification, segmentation, and a global searching solution, which are time-consuming and computationally complex. We propose a more efficient ROI extraction model for remote sensing images based on multiscale visual saliency analysis (MVS), implemented in the CIE L*a*b* color space, which is similar to visual perception of the human eye. We first extract the intensity, orientation, and color feature of the image using different methods: the visual attention mechanism is used to eliminate the intensity feature using a difference of Gaussian template; the integer wavelet transform is used to extract the orientation feature; and color information content analysis is used to obtain the color feature. Then, a new feature-competition method is proposed that addresses the different contributions of each feature map to calculate the weight of each feature image for combining them into the final saliency map. Qualitative and quantitative experimental results of the MVS model as compared with those of other models show that it is more effective and provides more accurate ROI extraction results with fewer holes inside the ROI.
Simple and effective segmentation algorithms are required for remote sensing images because of their mass data and complex texture features. An algorithm based on minimum class mean absolute deviation (MCMAD) is proposed. First, a two-dimensional (2-D) histogram is constructed by a median filter and gray process. Second, by using a diagonal projection, the 2-D histogram of remote sensing images is transformed into a one-dimensional (1-D) histogram to decrease the computational complexity. Finally, class mean absolute deviation of each threshold in the 1-D histogram is calculated and the threshold corresponding to the MCMAD is considered as the optimal segmentation threshold. To improve performance, we introduce spectral information into the MCMAD algorithm and the results of spectral bands are combined to get final segmentation results. Because most of the background used in our experiment is vegetation, we introduce a normalized difference vegetation index band into our algorithm and use the MCMAD algorithm on it. Experimental results show that our algorithms not only perform better for remote sensing images but also meet time requirements.
Remote sensing technology offers a very useful tool for understanding the earth. It brings us mass of information on the
earth and makes the view of human being broaden greatly. With the remote sensing technology developing, the spatial
resolution of image has been greatly upgraded. TIFF file format is a common format in graphics image processing.
GeoTIFF format is extended from TIFF. The GeoTIFF is widely used as a storage format of remote sensing. But the
format of TIFF and GeoTIFF is complex and they can’t be used efficiently under the windows operating system. In this
paper, an efficient file reading platform for high-resolution remote sensing image is proposed. This new design builds an
efficient reading platform of TIFF and GeoTIFF format file by image tiling and adding a cache in the middle of the
intermediate file. At the same time this design is a part of high resolution remote sensing image compression system. It
will have important theoretical and practical value for compression of the remote sensing images in the future.
The low computational complexity and high coding efficiency are the most significant requirements for image compression and transmission. Reversible biorthogonal integer wavelet transform (RB-IWT) supports the low computational complexity by lifting scheme (LS) and allows both lossy and lossless decoding using a single bitstream. However, RB-IWT degrades the performances and peak signal noise ratio (PSNR) of the image coding for image compression. In this paper, a new IWT-based compression scheme based on optimal RB-IWT and improved SPECK is presented. In this new algorithm, the scaling parameter of each subband is chosen for optimizing the transform coefficient. During coding, all image coefficients are encoding using simple, efficient quadtree partitioning method. This scheme is similar to the SPECK, but the new method uses a single quadtree partitioning instead of set partitioning and octave band partitioning of original SPECK, which reduces the coding complexity. Experiment results show that the new algorithm not only obtains low computational complexity, but also provides the peak signal-noise ratio (PSNR) performance of lossy coding to be comparable to the SPIHT algorithm using RB-IWT filters, and better than the SPECK algorithm. Additionally, the new algorithm supports both efficiently lossy and lossless compression using a single bitstream. This presented algorithm is valuable for future remote sensing image compression.
The IWT (Integer Wavelet Transform) can achieve genuine lossless image compression and allow both lossy and
lossless compression using a single bit-stream. However, using the IWT instead of the DWT (Discrete Wavelet
Transform) will degrade the performances of the lossy compression because the filter structure of IWT and the nonlinear
rounding operation. In this paper, a new integer wavelet decomposition scheme is proposed based on subband local
information statistic model for the medical images. The high frequency subbands can be decomposed again according to
the statistic results of the subband coefficient entropies. The results of several experiments for the medical images
presented in this paper demonstrate the importance subband information statistic in the integer wavelet decomposition.
Furthermore, this paper shows that appropriate subband local information statistic model improves the performance of
compression algorithm after the multilevel subband decomposition is performed. So we expect this idea is valuable for
future research on medical image coding.
Multiple Region of Interest (ROI) coding is important in applications where some parts of the image are of higher
importance than others and need to be encoded at higher quality than the background. The new image coding standard
JPEG2000 recommended the Maxshift method to complete ROI coding. However, the drawback of the Maxshift method
is that the coefficient bitplanes of all ROIs must be scaled with the same values, which cannot code ROIs according to
different degrees of interest. This paper describes a flexible multiple ROI coding scheme called BTShift (Bitplane Twice
Shift). The proposed method uses different bitplane scaling strategies between low frequency subbands and high
frequency subbands. Experimental results on several state-of-the-art images show that the new scheme does not only
support multiple ROI coding according to different degrees of interest, but also can encode the ROIs and background
without their shape information. Additionally, BTShift is compliant with the recommended Maxshift method. So we
hope the proposed method can be valuable for the remote sensing image compression and medical image coding in the
future.
Regions of Interest coding (ROI) is the important and convenient image coding technique and is supported by JPEG 2000. It enables a non-uniform distribution of the image quality between selected regions of interest and the background. This paper provides a novel ROI bitplants shift method based on compensation scheme called SHIFT-CS coding algorithm. The method firstly transmits the coding blocks of low-frequency important coefficients that are decoded. The residual coefficients in ROI mask can be transmitted on different bit-planes. When multiple regions of interest are encoded, every ROI has several bit-plane to be transmitted that can be resolved by the ROI's importance. In addition, different objectors can select different regions of interest by the low-quality reconstructive image according to different needs that realize to select ROI based on interactive net browser. The experiments show that SPIFT-CS method combines advantages of both ROI methods. Because the algorithm remain compliant with the MAXSHIFT decoding algorithm described in JPEG 2000 part 1, it is also simple and can be handled by any JPEG 2000 decoder.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.