Target track system design based on circular projection

Motivated by the recent success of supervised and weakly supervised common object discovery, in this paper, we move forward one step further to tackle common object discovery in a fully unsupervised way. Generally, object co-localization aims at simultaneously localizing objects of the same class across a group of images. Traditional object localization/detection usually trains specific object detectors which require bounding box annotations of object instances, or at least image-level labels to indicate the presence/absence of objects in an image. Given a collection of images without any annotations, our proposed fully unsupervised method is to simultaneously discover images that contain common objects and also localize common objects in corresponding images. Without requiring to know the total number of common objects, we formulate this unsupervised object discovery as a sub-graph mining problem from a weighted graph of object proposals, where nodes correspond to object proposals, and edges represent the similarities between neighbouring proposals. The positive images and common objects are jointly discovered by finding sub-graphs of strongly connected nodes, with each sub-graph capturing one object pattern. The optimization problem can be efficiently solved by our proposed maximal-flow-based algorithm. Instead of assuming that each image contains only one common object, our proposed solution can better address wild images where each image may contain multiple common objects or even no common object. Moreover, our proposed method can be easily tailored to the task of image retrieval in which the nodes correspond to the similarity between query and reference images. Extensive experiments on PASCAL VOC 2007 and Object Discovery data sets demonstrate that even without any supervision, our approach can discover/localize common objects of various classes in the presence of scale, view point, appearance variation, and partial occlusions. We also conduct broad experiments on image retrieval benchmarks, Holidays and Oxford5k data sets, to show that our proposed method, which considers both the similarity between query and reference images and also similarities among reference images, can help to improve the retrieval results significantly.

We present a novel approach to measuring distances between objects in images, suitable for information-rich object representations which simultaneously capture several properties in each image pixel. Multiple spatial fuzzy sets on the image domain, unified in a vector-valued fuzzy set, are used to model such representations. Distance between such sets is based on a novel point-to-set distance suitable for vector-valued fuzzy representations. The proposed set distance may be applied in, e.g., template matching and object classification, with an advantage that a number of object features are simultaneously considered. The distance measure is of linear time complexity w.r.t. the number of pixels in the image. We evaluate the performance of the proposed measure in template matching in presence of noise, as well as in object detection and classification in low resolution Transmission Electron Microscopy images.

A novel two stage template matching method for rotation and illumination invariance

Estimating the salient regions of an image plays a key role in scene analysis and image understanding. We can also apply saliency-based image processing techniques to image compression, evaluation, and effective searching methods. One of the most difficult problems is estimating the regions before recognizing what is in the image -- a problem that can be solved by accessing information of the low-level structures of objects in the image. This paper describes a method for estimating salient regions in images based on the distribution stability of local extrema of luminance during image blurring. Under blurring conditions, if an object's region has a more stable structure compared to another area, it must be more salient, so the saliency of these regions must be defined based on their stability for blurring. In the developed method, the local extrema of images are used to describe the complexity of the image's objects and background. Salient regions are estimated based on the stability of the local extrema for the blurring parameter. Experiments were conducted to compare the estimated result of salient regions and the psychophysical result obtained from the analysis of eye movement recordings. Results show that our method successfully extracts salient regions of natural images.

The X-ray scene image of a brain blood vessel of a rat can be shot by using SPring-8 that is a high-performance X-ray sensing system. However, the contrast of the scene image is not enough visually. Even if the contrast of a scene image is poor, a single image that has good contrast can be generated by accumulating all frames in the scene image. However, objects in the scene image must be fixed to use the method. This paper proposes a method to generate a single image with good contrast from X-ray scene images of brain blood vessel of a rat. The partial optimal frames that are useful for generating an output image with good contrast are selected in all frames and the selected frames are accumulated after optimal two-dimensional moving. The genetic algorithm is applied to optimize the selection and moving of frames. The experimental results show that the generated output image by the proposed method had good contrast, clearness and fewer noises in comparison with the generated image by the conventional method.

A novel method to recognize occluded objects using Markov model is proposed in this paper. In addition to Markov model having a high tolerance to noise, spatial distribution of features can be incorporated into Markov model in a natural and elegant way. Thus, high recognition accuracy can be achieved by the proposed method. More specifically, for each occluded object in the scene image, its translation, rotation and scale parameters can all be determined by our method even when it may have transformation parameters different from others or it may be duplicated in the scene image with transformation parameters different from each other. Moreover, the recognition process can be performed step by step to find out all of the objects in the scene image according to the confidence measure. Finally, the recognition process can be terminated automatically without knowing the number of objects included in the scene image since hypothesis verification and termination test are performed in our method. Actually, the solution of our method is useful for depth-search applications such as inspection of printed circuit board with multiple layers, underwater diving for searching objects, underground drilling for exploring mine, etc. The proposed method has been applied on two types of databases: puzzle and tool. The effectiveness and practicability of the proposed approach have been proven by various experimental results.

Rapid urbanization took place in the Pearl River Delta of south China since 1980. Although drastic land use change took place in very short interval within this area, hardly any research has been done on this phenomenon for lacking of available data. Remote sensing is presently the most favorable observation method for land use and land cover change (LUCC) researches. While located in the south of China, the Pearl River Delta suffers from heavy cloud cover for more than half of the year. This makes real-time LUCC monitoring and change detection almost impossible with commonly used optical remote sensing data. In this paper, the orbital highest resolution SAR (Synthetic Aperture Radar) data - Fine Mode Radarsat data was used for trail of short-term land use change detection. Three scenes of repeat-pass Radarsat data was collected over the study area. Although repeat-pass Radarsat enable continuous land use monitoring under all weather condition, images acquired during different time are inevitably affected by seasonal land cover change and variable environmental status such as air humidity and raining. Besides, some significant observation bias might be induced because of the platform and sensor instability. All these variations and instability made short-term land use change detection quite a perplex problem. In this paper, short-term land use change caused by human activity was considered as abnormal phenomena in both spatial and temporal domain in time series images. And a Density-based Anomaly Detection (DBAD) algorithm was designed to detect abnormally changed land parcels in time series Radarsat images. Firstly, totally 3 scenes of fine mode Radarsat images were collected in the study area from January 1st to May 3rd, 2006. Simply stacked temporal images reveal apparent backscattering variation between the three scenes of images, which mainly owes to the fast vegetable growth during the observation period. Then image segmentation was done on the multi-temporal Radarsat images and object features including mean value of backscattering coefficient (Mean), minimal value of backscattering (Min), homogeneity of gray level co-occurrence matrix (GLCMhomo) and dissimilarity of gray level co-occurrence matrix (GLCMdis) were extracted basing on segmented image objects. After that change- vector was constructed for each land objects. In the third step DBAD algorithm was applied to the change vector dataset to detect anomaly change in the 3 scenes of images. Finally field surveying data plus manual interpretation were used for validation. Comparing with object-based image regression method, DBAD results in better accuracy. Besides, data validation also shows that DBAD have better accuracy in both under-constructed area and newly built up area (error lower than 12%). While for built up area and some mixed used area, it gains relatively lower accuracy than other land types (from 10% to 28.57%). To conclude, short-term land use change in time series images could be defined as spatial and temporal anomaly in remote sensing images. By extending traditional anomaly detection to spatial-temporal anomaly detection, land use change caused by human activity could be effectively detected during short time intervals. The algorithm DBAD focus only on the density of change vectors in feature space, which is independent of the amplitude and direction of change vectors. This enable DBAD effectively discriminate temporal image variation caused by observation system, environment or seasonal land cover change, especially in vegetation and cultivated area which changed remarkably during the observation period, from land use change caused by human activities. This helps to decrease the false alarming in short-term change detection. © 2008 SPIE.

Template matching is used for many applications in image processing. Cross Correlation is the basic statistical approach to image registration. It is used for template matching or pattern recognition. Template can be considered a sub-image from the reference image, and the image can be considered as a sensed image. The objective is to establish the correspondence between the reference image and sensed image. It gives the measure of the degree of similarity between an image and template. This paper describes medical image registration by template matching based on Normalized Cross-Correlation (NCC) using Cauchy-Schwartz inequality. The algorithm for template matching using NCC is implemented in MATLAB. The algorithm does the template matching and uses the Cauchy-Schwartz’s inequality to simplify the procedure. The developed algorithm is robust for similarity measure. An experimental result with medical images registration with noise and without noise is shown in the results section.

Template matching is a basic algorithm for image processing, and real-time is a crucial requirement of object tracking. For real-time tracking, a fast template matching algorithm based on grey prediction is presented, where computation cost can be reduced dramatically by minimizing search range. First, location of the tracked object in the current image is estimated by Grey Model (GM). GM(1,1), which is the basic model of grey prediction, can use some known information to foretell the location. Second, the precise position of the object in the frame is computed by template matching. Herein, Sequential Similarity Detection Algorithm (SSDA) with a self-adaptive threshold is employed to obtain the matching position in the neighborhood of the predicted location. The role of threshold in SSDA is important, as a proper threshold can make template matching fast and accurate. Moreover, a practical weighted strategy is utilized to handle scale and rotation changes of the object, as well as illumination changes. The experimental results show the superior performance of the proposed algorithm over the conventional full-search method, especially in terms of executive time.

In digital image processing, template matching is a process to determine the location of sub image inside an image. The sub image, which is called template, usually has similarity with a part of the image. The template can be in different size, color or form. Template matching is famously used in image registration and object recognition. In this paper, we focus on the performance of the Sum of Squared Differences (SSD) and Normalized Cross Correlation (NCC)as the techniques that used in image registration for matching the template with an image. This experiment is aiming to compare the ability of both techniques in term of quality of output image as well as the time taken in execution process. Furthermore, it also to test the effect of template image to output image when there is noise and rotation. This paper provides an explanation about the concept of similarity measure techniques and how the algorithms are implementing in image registration process. Finally, the performance of these methods is tested by making comparison based on the value of correlation coefficient that produced from different image templates.

PurposeContent‐based image retrieval (CBIR) technologies offer many advantages over purely text‐based image search. However, one of the drawbacks associated with CBIR is the increased computational cost arising from tasks such as image processing, feature extraction, image classification, and object detection and recognition. Consequently CBIR systems have suffered from a lack of scalability, which has greatly hampered their adoption for real‐world public and commercial image search. At the same time, paradigms for large‐scale heterogeneous distributed computing such as grid computing, cloud computing, and utility‐based computing are gaining traction as a way of providing more scalable and efficient solutions to large‐scale computing tasks.Design/methodology/approachThis paper presents an approach in which a large distributed processing grid has been used to apply a range of CBIR methods to a substantial number of images. By massively distributing the required computational task across thousands of grid nodes, very high through‐put has been achieved at relatively low overheads.FindingsThis has allowed one to analyse and index about 25 million high resolution images thus far, while using just two servers for storage and job submission. The CBIR system was developed by Imense Ltd and is based on automated analysis and recognition of image content using a semantic ontology. It features a range of image‐processing and analysis modules, including image segmentation, region classification, scene analysis, object detection, and face recognition methods.Originality/valueIn the case of content‐based image analysis, the primary performance criterion is the overall through‐put achieved by the system in terms of the number of images that can be processed over a given time frame, irrespective of the time taken to process any given image. As such, grid processing has great potential for massively parallel content‐based image retrieval and other tasks with similar performance requirements.

Knowledge of the number and distribution of oil palm trees during the crop cycle is vital for sustainable management and predicting yields. The accuracy of the conventional image processing method is limited for the hand-crafted feature extraction method and the overfitting problem occurs due to the insufficient dataset. We propose a modification of the Faster Region-based Convolutional Neural Network (FRCNN) for palm tree detection to reduce the overfitting problem and improve the detection accuracy. The enhanced FRCNN (EFRCNN) leads to improved performance for detecting objects (in the same image) when they are of multiple sizes by using a feature concatenation method. Transfer learning based on a ResNet50 model is used to extract the features of the input image. High-resolution images of oil palm trees from a drone are used to form the data set, containing mature, young, and mixed oil palm tree regions. We train and test the EFRCNN, the FRCNN, a CNN used recently for oil palm image detection, and two standard methods, namely, the support vector machine (SVM) and template matching (TM). The results reveal an overall accuracy of ≥96.8% for the EFRCNN on the three test sets. The accuracy is higher than the CNN and FRCNN and substantially higher than SVM and TM. For large-scale plantations, the accuracy improvement is significant. This research provides a method for automatically counting the oil palm trees in large-scale plantations.

This paper describes an approach for finding a signboard along with the camera viewing direction in a gray scene image and rectifying a distorted signboard image taken from a nonorthogonal viewing direction. First, we use heuristics of both characters and character lines to discriminate characters from other objects in a gray scene image. Then, we attempt to compute the orientation of the camera relative to a detected signboard from a single view. Finally, we rectify the distortion of signboard images that are viewed at an angle. The only restriction for viewing direction estimation and distortion rectification is that a detected signboard is rectangular. We evaluate our methods on both simulation images and real scene images. Experimental results demonstrate the effectiveness of the proposed methods.

Image template matching refers to the technique of locating a given reference image over a source image such that they are the most similar. It is a fundamental mission in the field of visual target recognition. In general, there are two critical aspects of a template matching scheme. One is similarity measurement and the other is best-match location search. In this work, we choose the well-known normalized cross correlation model as a similarity criterion. The searching procedure for the best-match location is carried out through an internal-feedback artificial bee colony (IF-ABC) algorithm. IF-ABC algorithm is highlighted by its effort to fight against premature convergence. This purpose is achieved through discarding the conventional roulette selection procedure in the ABC algorithm so as to provide each employed bee an equal chance to be followed by the onlooker bees in the local search phase. Besides that, we also suggest efficiently utilizing the internal convergence states as feedback guidance for searching intensity in the subsequent cycles of iteration. We have investigated four ideal template matching cases as well as four actual cases using different searching algorithms. Our simulation results show that the IF-ABC algorithm is more effective and robust for this template matching mission than the conventional ABC and two state-of-the-art modified ABC algorithms do.

5 - High-level feature extraction: fixed shape matching