Search

Image classification plays a fundamental role in hyperspectral image (HSI) analysis. Since the mixed pixels of the urban areas are generally more complex than other areas, the following two problems remain to be considered while dealing with urban HSI classification: 1) due to the fact that the spectral feature of different mixed pixels varies greatly in the same class, HSI classification of urban area is susceptible to the representativeness of the training samples and 2) since the urban area is densely packed with objects of different size, the comprehensive use of the spatial and spectral features to classify the objects is a difficult problem. To tackle these problems, HSI classification based on multiscale spectral-spatial deformable network (S^2-DNet) is proposed. First, a k-means clustering method is adopted to cluster spectrum of each class, and representative samples are selected from the spectrum after clustering to reduce the impact of intraclass variation. Second, a spectral-spatial joint network is designed to extract the low-level features, including spectral features and spatial features. Third, the deformable network is introduced to extract high-level features of the object. Experimental results demonstrated that the proposed method outperformed the state-of-the-art methods on two widely used HSI data sets.

It is observed that the great spectral variation in the same hyperspectral image (HSI) pixel class often leads to misclassification. To solve this problem, we have proposed an automatic clustering-based two-branch convolutional neural network (CNN): first, to reduce the intraclass spectral variation, the HSI pixels are automatically subdivided into smaller classes by clustering; second, in order to suppress the interference of spectral amplitude variation, the SincNet is introduced to capture the spectral pattern by giving more weight to the spectral shape; third, the DS-CNN with double directional strip convolution kernel is designed to extract spatial feature, so that specific contextual interactional features can be collected, especially in strip-shaped field-like roads and farmlands; finally, the spectral and spatial features extracted by the two branches are fused at fully connected layer to obtain an accurate classification. Extensive experiments demonstrated that the proposed method can obtain better classification performance than the state-of-the-art methods.

Vehicle detection is a significant and challenging task in aerial remote sensing applications. Most existing methods detect vehicles with regular rectangle boxes and fail to offer the orientation of vehicles. However, the orientation information is crucial for several practical applications, such as the trajectory and motion estimation of vehicles. In this paper, we propose a novel deep network, called rotatable region-based residual network (R$^3$-Net), to detect multi-oriented vehicles in aerial images and videos. More specially, R$^3$-Net is utilized to generate rotatable rectangular target boxes in a half coordinate system. First, we use a rotatable region proposal network (R-RPN) to generate rotatable region of interests (R-RoIs) from feature maps produced by a deep convolutional neural network. Here, a proposed batch averaging rotatable anchor (BAR anchor) strategy is applied to initialize the shape of vehicle candidates. Next, we propose a rotatable detection network (R-DN) for the final classification and regression of the R-RoIs. In R-DN, a novel rotatable position sensitive pooling (R-PS pooling) is designed to keep the position and orientation information simultaneously while downsampling the feature maps of R-RoIs. In our model, R-RPN and R-DN can be trained jointly. We test our network on two open vehicle detection image datasets, namely DLR 3K Munich Dataset and VEDAI Dataset, demonstrating the high precision and robustness of our method. In addition, further experiments on aerial videos show the good generalization capability of the proposed method and its potential for vehicle tracking in aerial videos. The demo video is available at https://youtu.be/xCYD-tYudN0.

In this study, we employ our developed instrument to obtain high-throughput multi-angle single-particle polarization scattering signals. Based on experimental results of a variety of samples with different chemical composition, particle size, morphology, and microstructure, we trained a deep convolutional network to identify the polarization signal characteristics during aerosol scattering processes, and then investigate the feasibility of multi-dimensional polarization characterization applied in the online and real-time fine and accurate aerosol recognition. Our model shows a high classification accuracy rate (>98%) and can achieve aerosol recognition at a very low proportion (

Aerosol particle classification are of great importance in many cases. But currently there is almost no universal method to deal with this task. In this study, we develop a one-dimensional convolution neural network taking multi-angle polarization time series signal of single suspended particle as input to identify its category. We train the network and reach quite high accuracy on a large dataset which contains signals of multi-kinds particles such as carbon black, PSL, dust and water-soluble salts. This method gives us a new way of looking deeper into single suspended particles in the air and the knowledge learned in this task may be able to be transferred to deal with tasks of recognizing more kinds or more complex aerosol particles such as bioaerosol or even airborne pathogen.

We propose a hybrid method combining the deep learning technique and the ratio enhancement (RE) method for pansharpening. The intuition behind is to utilize the deep learning technique to synthesize a panchromatic (PAN) image for the RE method to reduce the spectral distortion while keeping the spatial details. The method consists of two stages. First, the CNN synthesizer is optimized to generate the downsampled PAN image to guarantee the network have a good initialization. Second, CNN is integrated into the RE method and supervised by the ground truth multi-spectral (MS) to produce an ideal synthesized PAN for the RE method. We conduct experiments on various datasets and compare with widely used methods to demonstrate the superiority of the proposed method.

Ship detection from thermal remote sensing imagery is a challenging task because of cluttered scenes and variable appearances of ships. In this letter, we propose a novel detection algorithm named region-based deep forest (RDF) toward overcoming these existing issues. The RDF consists of a simple region proposal network and a deep forest ensemble. The region proposal network trained over gradient features robustly generates a small number of candidates that precisely cover ship targets in various backgrounds. The deep forest ensemble adaptively learns features from remote sensing data and discriminates real ships from region proposals efficiently. The training process of deep forest ensemble is efficient and users can control training cost according to computational resource available. Experimental results on numerous thermal satellite images demonstrate the superior performance of our method compared with state-of-the-art methods.

Automatic ship detection from optical satellite imagery is a challenging task due to cluttered scenes and variability in ship sizes. This letter proposes a detection algorithm based on saliency segmentation and the local binary pattern (LBP) descriptor combined with ship structure. First, we present a novel saliency segmentation framework with flexible integration of multiple visual cues to extract candidate regions from different sea surfaces. Then, simple shape analysis is adopted to eliminate obviously false targets. Finally, a structure-LBP feature that characterizes the inherent topology structure of ships is applied to discriminate true ship targets. Experimental results on numerous panchromatic satellite images validate that our proposed scheme outperforms other state-of-the-art methods in terms of both detection time and detection accuracy.

This letter presents a novel change detection method for multitemporal synthetic aperture radar images based on PCANet. This method exploits representative neighborhood features from each pixel using PCA filters as convolutional filters. Thus, the proposed method is more robust to the speckle noise and can generate change maps with less noise spots. Given two multitemporal images, Gabor wavelets and fuzzy $c$ -means are utilized to select interested pixels that have high probability of being changed or unchanged. Then, new image patches centered at interested pixels are generated and a PCANet model is trained using these patches. Finally, pixels in the multitemporal images are classified by the trained PCANet model. The PCANet classification result and the preclassification result are combined to form the final change map. The experimental results obtained on three real SAR image data sets confirm the effectiveness of the proposed method.

The existing relative radiometric normalization methods are insufficient to define the invariant pixels automatically, and the conventional methods do not perform well when the multitemporal images contain a lot of changes. Two types of changes should be particularly considered: one is caused by significant spectral differences due to change of ground objects, and the other is the pixels in the regions of misalignment caused by displacement due to differences in acquisition view angles and geometrical distortions. To automatically extract invariant pixels and reduce the influence of the changes, a hierarchical regression method is proposed to reduce the radiation difference for multitemporal images, which consists of extraction of the pseudo-invariant features (PIFs) and optimization of normalization parameters. A weighted regression based on spectral difference is proposed to automatically extract the PIFs, which can also suppress the negative effect of the first type of changes. In addition, a robust regression with gradient dependence is performed on the extracted PIFs to build the final relationship between the target image and the reference image, which can be robust for the second type of changes. Experimental results demonstrate that the proposed method has a better performance to normalize the target image.

Hyperspectral image classification has been acknowledged as the fundamental and challenging task of hyperspectral data processing. The abundance of spectral and spatial information has provided great opportunities to effectively characterize and identify ground materials. In this paper, we propose a spectral and spatial classification framework for hyperspectral images based on Random Multi-Graphs (RMGs). The RMG is a graph-based ensemble learning method, which is rarely considered in hyperspectral image classification. It is empirically verified that the semi-supervised RMG deals well with small sample setting problems. This kind of problem is very common in hyperspectral image applications. In the proposed method, spatial features are extracted based on linear prediction error analysis and local binary patterns; spatial features and spectral features are then stacked into high dimensional vectors. The high dimensional vectors are fed into the RMG for classification. By randomly selecting a subset of features to create a graph, the proposed method can achieve excellent classification performance. The experiments on three real hyperspectral datasets have demonstrated that the proposed method exhibits better performance than several closely related methods.

Sea-land segmentation is a key procedure of ship detection. However, most of the existing sea-land segmentation methods are designed for on-ground ship detection. Consequently, the computational costs of these methods are too expensive to be implemented in on-orbit platforms. To tackle this problem, first, a shoreline database is built according to the global geographic information systems; then, the sea-land segmentation method is proposed by utilizing the corresponding coastline data, which is obtained from the database according to the latitude and longitude of the candidate area. Compared to the existing sea-land segmentation methods, this method is more suitable for on-orbit platforms, because of the less computational costs, higher efficient, and acceptable robustness. The experimental results based on raw panchromatic images demonstrated that the proposed method had a good performance in sea-land segmentation for on-orbit processing.

Accurate interpretation of high spatial resolution multispectral (MS) imagery relies on the extraction and fusion of information obtained from both spectral and spatial domains. Feature extraction from one or several fixed windows uses inaccurate description of pixel contexts and produces blurred object boundaries and low classification accuracy. In order to accurately characterize the spatial context properties of pixels, this paper presents a hierarchical-segmentation-based classification system. The system consists of two main modules: 1) hierarchical segmentation and 2) context-based classification. The segmentation module involves an optimization procedure to prevent undersegmentation of the land objects of interest and a scale selection procedure to find the most representative segmentation layers for modeling pixel contexts. The classification module couples a context-driven multilevel feature extraction methodology with a support vector machine classifier to get classification result. The proposed system is validated on three high spatial resolution MS data sets. Compared with state-of-the-art classification methods based on the similar concept, the proposed method demonstrates superior performance on both the classification accuracy and the quality of classification maps.

The synthesis of low-resolution panchromatic (Pan) image is a critical step of ratio enhancement (RE) and component substitution (CS) pansharpening methods. The two types of methods assume a linear relation between Pan and multispectral (MS) images. However, due to the nonlinear spectral response of satellite sensors, the qualified low-resolution Pan image cannot be well approximated by a weighted summation of MS bands. Therefore, in some local areas, significant gray value difference exists between a synthetic Pan image and a high-resolution Pan image. To tackle this problem, the pixels of Pan and MS images are divided into several classes by k-means algorithm, and then multiple regression is used to calculate summation weights on each group of pixels. Experimental results demonstrate that the proposed technique can provide significant improvements on reducing color distortion.

Due to the difference of “mean information” between substitution component and substituted component, spectral distortion often occurs in component substitution (CS) pansharpening. In this paper, a data fitting scheme is adopted to improve spectral quality in image fusion based on well-established CS approach. A generalized CS framework that is capable of modeling any CS image fusion method is also presented. In this framework, instead of injecting detail information of panchromatic (Pan) image into substituted component, the data fitting strategy is designed to adjust the mean information of Pan image in the construction of substitution component. The data fitting scheme involves two matrix subtractions and one matrix convolution. It is fast in implementation and is effective to avoid the spectral distortion problem. Experimental results on a large number of Pan and multispectral images show that the improved CS methods have good performance on the spatial and spectral fidelity. Moreover, experiments carried out on large-size images also show an excellent running time performance of the proposed methods.

Due to the fact that WorldView-2 (WV2) has a small time lag while acquiring images from panchromatic (PAN) and two multispectral (MS1 and MS2) sensors, a moving vehicle is located at different positions in three image bands. Consequently, such displacement can be utilized to identify moving vehicles, and vehicle information, such as speed and direction can be estimated. In this paper, we focus on moving vehicle detection according to the displacement information and present a novel processing chain. The vehicle locations are extracted by an improved morphological detector based on the vehicle’s shape properties. To make better use of the time lag between MS1 and MS2, a band selection process is performed by both visual inspection and quantitative analysis. Moreover, three spectral-neighbor band pairs, which have a major contribution to vehicle identification, are selected. In addition, we improve the spatial and spectral analysis method by incorporating local ERGAS index analysis (ERGAS-SNS) to identify moving vehicles. The experimental results on WV2 images showed that the correctness, completeness and quality rates of the proposed method were about 94%, 91% and 86%, respectively. Thus, the proposed method has good performance for moving vehicle detection and information extraction.

Pansharpening is a process of fusing the spatial detail of a high-resolution panchromatic (Pan) image and the spectral information of a low-resolution multispectral (MS) image to generate a high-resolution MS image. With the fast development of high-resolution satellites, pansharpening techniques are increasingly used worldwide. Even though pansharpening research has been done for decades and numerous techniques have been developed, research efforts are still continuing to further improve the techniques. The most common problems encountered by pansharpening algorithms are colour distortion and spatial artefacts. This article provides a review on recent advances in solving or reducing the problems of colour distortion and spatial artefacts. On the other hand, miss-registration and object size difference between Pan and MS images are also two common key problems in pansharpening applications. Very few publications have discussed these issues. Therefore, this article also addresses these two key problems and...

Due to high rate of false match and expensive computation cost, the existing scale-invariant feature transform (SIFT) operators are not efficient to register two optical satellite images that have a great spatial resolution difference. Some scale restriction schemes were proposed to reduce the false match of SIFT keypoints and computational cost. However, it is observed that many keypoints are still not correctly matched. This problem often leads to the failure of automatic registration in industrial applications. To solve this problem, the images being registered are normalized to adjust the scale of blob-like structures and to preclude useless blob-like structures. Then blob-like structures are classified according to their physical sizes, and keypoint matching is restricted to matching for the blob-like structures having the same physical sizes. The scale normalization and size classification significantly improve the correct match rate as well as computational cost. Experiments on two pairs of satelli...

Automatic ship detection in high-resolution optical satellite images with various sea surfaces is a challenging task. In this letter, we propose a novel detection method based on sea surface analysis to solve this problem. The proposed method first analyzes whether the sea surface is homogeneous or not by using two new features. Then, a novel linear function combining pixel and region characteristics is employed to select ship candidates. Finally, Compactness and Length-width ratio are adopted to remove false alarms. Specifically, based on the sea surface analysis, the proposed method cannot only efficiently block out no-candidate regions to reduce computational time, but also automatically assign weights for candidate selection function to optimize the detection performance. Experimental results on real panchromatic satellite images demonstrate the detection accuracy and computational efficiency of the proposed method.

Due to the different imaging modalities and acquisition time, keypoint-based registration methods often suffer from false matches of keypoints while utilizing to register the optical remote sensing images from multi-sensors. In this paper, we proposed a novel method based on Line-Point Invariant for the multi-sensor image registration. First, the line segments of the images are extracted, and then the salient line segments are detected depending upon the adaptive confidence. Subsequently, conjugate salient lines between the two images are identified as the registration primitives by the probability relaxation labelling approach. Second, we obtain the SIFT keypoints of the images and establish the matches of the keypoints based on the Line-Point Invariant via dual matching. Consequently, false keypoint matches are greatly reduced and the correct match rate is significantly enhanced. The experiments conducted on various multi-sensor images demonstrate the effectiveness of the proposed method.

Automatic ship detection from optical satellite images remains a tough task. In this paper, a novel method of ship detection from optical satellites is proposed by analyzing the ship distribution probability. First, an anomaly detection model is constructed by the sea cluster histogram model; then, the ship distribution based on the ship safety navigational criterion is analyzed to obtain the ship candidates, and obvious non-ship objects are removed by the area properties from ship candidates; finally, a structural continuity descriptor is designed to remove false alarms from the ship candidates. Experiments on numerous satellite images from panchromatic and one band within multispectral sensors are conducted. The results verified that the proposed method outperforms existing methods in both effectiveness and efficiency.

This letter presents a novel coarse-to-fine level set method for contour extraction in optical satellite images. To distinguish objects from a background, the undecimated wavelet transform is firstly adopted to extract image features, and a homogeneity metric is defined to measure the variation of the features inside and outside contours. In addition, the weight distribution ratio is proposed to adaptively tune the relative weight of the features. Based on the homogeneity metric and the weight distribution ratio, a novel energy functional is developed to model a contour extraction problem, and in order to reduce the computation burden, a coarse-to-fine scheme is applied to progressively extract contours in finer scale, during which a contour position constraint is introduced to limit contours evolving in a small space around the candidate contours extracted in coarser scale. Extensive experiments have been carried out on optical satellite images to validate the proposed method.

Automatic ship detection from high-resolution optical satellite images has attracted great interest in the wide applications of maritime security and traffic control. However, most of the popular methods have much difficulty in extracting targets without false alarms due to the variable appearances of ships and complicated background. In this paper, we propose a ship detection approach based on visual search mechanism to solve this problem. First, salient regions are extracted by a global contrast model fast and easily. Second, geometric properties and neighborhood similarity of targets are used for discriminating the ship candidates with ambiguous appearance effectively. Furthermore, we utilize the SVM algorithm to classify each image as including target(s) or not according to the LBP feature of each ship candidate. Extensive experiments validate our proposed scheme outperforms the state-of-the-art methods in terms of detection time and accuracy.

Pansharpening technique is very important for many remote sensing applications. Many fusion algorithms have been proposed to pan-sharpen multispectral (MS) images. However, there are still some spatial or spectral distortion problems in fusion result. There are two major reasons: First, panchromatic (PAN) image contains some interference spectrum information similar to MS image which may cause color distortion in the fusion result. Second, MS image has some interference spatial information approximates PAN image which may cause spatial artifacts. It is difficult to simultaneous eliminate the interference information from PAN and MS images. To solve the above problems, the paper presents an improved pan-sharpen algorithm which integrates the advantages of the ratio enhancement method and Gaussian-fitting. The high-frequency information of each ith band of MS image and the low-frequency information of PAN image are extracted by Gaussian-fitting, and the information is synthesized into a group of low-resolution PAN images. Finally, each ith band of MS image is pan-sharpened by a ratio enhancement, in which the ratio is obtained by image division between the PAN image and the ith synthesized low-resolution PAN image. Extensive experiments have been implemented on WorldView-2 images. Visual comparison and quantitative analysis demonstrated that the proposed method can achieve good performance in spatial and spectral fidelity.

The change of buildings is one of the most valuable information in the monitoring of land use for urban areas. Change detection technique based on multitemporal remote sensing image is an effective approach to obtain information of feature change. However, with the continuous improvement of resolution in remote sensing image, conventional change detection methods have much difficulty in exactly extracting building changes. One difficulty is the displacement for buildings between the Multitemporal remote sensing images due to the different view angles of sensors, another difficulty is the shadows, the above-mentioned difficulties are highlighted in the high-resolution remotely sensed Images. In this paper, a novel method for the detection of building changes from high-resolution images in urban areas is proposed, the candidate changed areas are obtained base on the spectral difference, and then a semantic dependency relation is integrated by a morphological building index technique and a shadow detection method to identify the real changes. The proposed method is evaluated with a pair of QuickBird images of Qingdao City, China. Experimental results demonstrate that the proposed method have a better performance to extract the building changes.

The goal of ratio enhancement for hyperspectral (HS) image pansharpening is to obtain an enhancement ratio between a simulated low-resolution panchromatic (Pan) image and an original high-resolution Pan image. However, the simulated low-resolution Pan image often suffers from gray-level distortion. To solve these problems, the original HS bands are synthesized to a smaller number of reduced HS bands, then the pixels of Pan and HS images are divided into different groups according to the linearity between the Pan band and the reduced bands. For each pixel group, a nonnegative least-squares algorithm is utilized to calculate the weights of reduced HS bands, so that the simulated Pan image is obtained by weighted summation of reduced HS bands. Finally, the HS image is sharpened by a ratio enhancement. The experiments demonstrated that the proposed method had a good performance on fusion quality.

An improved mean shift segmentation method featuring adaptive parameter selection is presented in this paper. We associate the bandwidths and weight for each point in a spatial-range feature space with boundary information in an image plane. Varying weight and bandwidth for each pixel are assigned according to a boundary map, which is obtained by learning multiple edge cues. We consider two groups of edge cues and two regressing modules, approach the cue combination as a supervised learning problem from the ground truth data (manually sketched boundary maps). From our preliminary results, the provided method can combine the top-down information got from regression models with the mean shift process and constrain over-clustering of pixels belonging different land objects.

Ratio enhancement (RE) methods are typical fusion techniques that have been widely used in multispectral image pansharpening. The construction of a low-resolution Pan image is a crucial step of RE methods. However, it is observed that the existing RE methods cause serious color distortion in HS image pansharpening. This problem arises from the ideal assumptions of Pan and HS images: (i) the relation that exists between Pan and HS bands is linear; (ii) regional variations of Pan and HS images do not have influence on pansharpening results; (iii) the multiple linear regression often fails to obtain the accurate weights of HS bands while the number of HS bands is more than 10. To solve this problem, the original HS bands are reduced to a small amount of synthesized HS bands by an average of neighbor bands, and then the pixels of Pan and HS images are divided into different groups. For each pixel group, the multiple linear regression between Pan and HS bands is utilized to calculate the weights of synthesized HS bands to synthesize a qualified low-resolution Pan image. Finally, the HS image is fused by a ratio enhancement, in which the ratio is obtained by image division between the Pan image and the synthesized Pan image. The experiments on EO-1 data sets demonstrated that the proposed method had a good performance on fusion quality, and outperformed the compared methods.

This paper presents a hierarchical structure feature analysis technique, which is robust to registration noise for change detection in multi-temporal very high spatial resolution (VHR) remote sensing images. To suppress the registration noise in “difference images”, the proposed technique extracts the structural features of the same local area from multi-temporal images and then analyzes the similarity of these features. Specifically, this method consist two steps: first, the structural features is employed to obtain gross suppression of registration noise, so that the noise from the regions that have high similarity is eliminated; second, the remained noise is further suppressed using the local features with more spatial details. Experimental results obtained from multi-temporal remote sensing images demonstrated the effectiveness of the proposed approach.

Handwriting recognition and document processing are formidable tasks, and the recognition of the different data fields such as courtesy amount, legal amount, date, payee and signature, are advancing at different paces. After examining the problems involved and the facts, this paper focuses on one of the most challenging parts of a cheque recognition system, i.e., the segmentation and recognition of the date written on the cheques. It mimics humans in segmenting a date image, and two separators are used to detect the `Year', `Day' and `Month' zones based on shape and spatial features. Experiments have been conducted on a test set of 809 cheques with diverse colour designs and languages, with encouraging results. Apart from the above, this paper also analyses the outstanding issues in this growing field, and attempts to provide some pointers and solutions.

Aircraft automatic detection from very high-resolution (VHR) images plays an important role in a wide variety of applications. This paper proposes a novel detector for aircraft detection from very high-resolution (VHR) remote sensing images. To accurately distinguish aircrafts from background, a circle-frequency filter (CF-filter) is used to extract the candidate locations of aircrafts from a large size image. A multi-level feature model is then employed to represent both local appearance and spatial layout of aircrafts by means of Robust Hue Descriptor and Histogram of Oriented Gradients. The experimental results demonstrate the superior performance of the proposed method.

Change detection is of high practical value to hazard assessment, crop growth monitoring, and urban sprawl detection. A synthetic aperture radar (SAR) image is the ideal information source for performing change detection since it is independent of atmospheric and sunlight conditions. Existing SAR image change detection methods usually generate a difference image (DI) first and use clustering methods to classify the pixels of DI into changed class and unchanged class. Some useful information may get lost in the DI generation process. This paper proposed an SAR image change detection method based on neighborhood-based ratio (NR) and extreme learning machine (ELM). NR operator is utilized for obtaining some interested pixels that have high probability of being changed or unchanged. Then, image patches centered at these pixels are generated, and ELM is employed to train a model by using these patches. Finally, pixels in both original SAR images are classified by the pretrained ELM model. The preclassification result and the ELM classification result are combined to form the final change map. The experimental results obtained on three real SAR image datasets and one simulated dataset show that the proposed method is robust to speckle noise and is effective to detect change information among multitemporal SAR images.

The anisotropic scale space (ASS) is often used to enhance the performance of a scale-invariant feature transform (SIFT) algorithm in the registration of synthetic aperture radar (SAR) images. The existing ASS-based methods usually suffer from unstable keypoints and false matches, since the anisotropic diffusion filtering has limitations in reducing the speckle noise from SAR images while building the ASS image representation. We proposed a speckle reducing SIFT match method to obtain stable keypoints and acquire precise matches for the SAR image registration. First, the keypoints are detected in a speckle reducing anisotropic scale space constructed by the speckle reducing anisotropic diffusion, so that speckle noise is greatly reduced and prominent structures of the images are preserved, consequently the stable keypoints can be derived. Next, the probabilistic relaxation labeling approach is employed to establish the matches of the keypoints then the correct match rate of the keypoints is significantly increased. Experiments conducted on simulated speckled images and real SAR images demonstrate the effectiveness of the proposed method.

In this paper, we propose a new method to extract object contour in a given gray-level image, whose foreground and background are statistically homogeneous and different. Firstly, the image for contour extraction is decomposed by discrete wavelet transform, and the high-pass and low-pass components are used to form intensity and texture energy respectively. Secondly, a minimal partition function, which combines intensity, texture and contour length energy, is made to model the contour extraction problem. Finally, the model is formulated in terms of level set function to obtain a numerical solution. Experiments have been performed on synthetic and remote-sensing images, and the results demonstrated that our method can adaptively use intensity and texture information to accurately extract object contour.

In this paper, a novel algorithm based on phase correlation and gray-scale projection is proposed to estimate the global motion vectors between each two consecutive image frames in jitter videos. In the algorithm, phase correlation is used firstly to estimate the approximate global motion vectors. Then the gray-scale projection algorithm based on the results of the phase correlation is used to generate the accurate global motion vectors. This algorithm has the characteristic of low computation load, high precision, wide range and strong anti-jamming. The experiments results demonstrate that the proposed algorithm can generate accurate global motion vectors with errors less than one pixel in real time for the jitter video sequence with the size of 704×576.

Existing segmentation methods require manual interventions to optimally extract objects from cluttered background, so that they can hardly work well in automated surveillance systems. In order to automatically extract aircrafts from very high-resolution images, we proposed a segmentation method that combines bottom-up and top-down cues. Three essential principles from local contrast, global contrast, and center bias are involved to compute bottom-up cue. In addition, top-down cue is computed by incorporating aircraft shape priors, and it is achieved by training a classifier from a rich set of visual features. Iterative operations and adaptive fitting are designed to get refined results. Experimental results demonstrated that the proposed method can provide significant improvements on the segmentation accuracy.

This paper describes a system being developed to recognizedate information handwritten on Canadian bankcheques. A segmentation based strategy is adopted in thissystem. In order to achieve high performances in terms ofefficiency and reliability, a knowledge-based module is proposedfor the date segmentation and a cursive month wordrecognition module is implemented based on a combinationof classifiers. The interaction between the segmentation andrecognition stages is properly established by using multi-hypothesesgeneration and evaluation modules. As a result,promising performance is obtained on a test set from a real-lifestandard cheque database.

Presents an approach to differentiating handwritten alphabetic words from numeric strings, which is motivated by an application in a real-life bank cheque processing system. Neural networks are first considered with the focus on effective feature extraction and architecture selection. The combination of networks is then implemented to achieve highly reliable and accurate results.

Segmenting handwritten date fields on bank cheque images into three subimages corresponding to the day, month and year is the first and critical step of our date recognition system. The paper describes a knowledge-based segmentation system, which introduces different kinds of knowledge at different segmentation stages to improve the performance. The knowledge includes information on the writing style, syntactic and semantic constraints, etc. Results have shown that the system is very effective compared with a previous structural feature based method.

The last frontiers of handwriting recognition are considered to have started in the last decade of the second millennium. The paper summarizes (a) the nature of the problem of handwriting recognition, (b) the state of the art of handwriting recognition at the turn of the new millennium, and (c) the results of CENPARMI researchers in automatic recognition of handwritten digits, touching numerals, cursive scripts, and dates formed by a mixture of the former 3 categories. Wherever possible, comparable results have been tabulated according to techniques used, databases, and performance. Aspects related to human generation and perception of handwriting are discussed. The extraction and usage of human knowledge, and their incorporation into handwriting recognition systems are presented. Challenges, aims, trends, efforts and possible rewards, and suggestions for future investigations are also included.

This article describes a system being developed to segment handwritten date fields on bank cheque images. The handwritten information extracted from the date zone is segmented into day, month and year, and a hypothesis is also made on the writing style of the month (in word or digits). The system has been implemented and tested on cheque images. Subsequent modifications have also been designed and implemented to include contextual information in the determination of segmentation points. Results have shown that the system is effeective; with continuing improvements, the system is expected to be a useful component for processing the date written on cheques.