Contour Pixels Research Articles

Persian Handwritten Digit Recognition using Support Vector Machines

In this paper, appropriate features set based on Discrete Fourier Transform coefficients and the box approach have been proposed to achieve higher recognition accuracy, decreasing the features set dimensions and recognition time of Persian numerals. In classification phase, support vector machine (SVM) has been employed as the classifier. Feature sets consists of 154 dimensions, which are the Fourier coefficients in the contour pixels of input image, average angle and distance pixels which are equal to one in each box the box approach. The scheme has been evaluated on 80,000 handwritten samples of Persian numerals. Using 60,000 samples for training, scheme was tested on other 20,000 samples and 98.84% correct recognition rate was obtained.

Creeping Contours: A Multilabel Image Segmentation Method for Extracting Boundary Surfaces of Parts in Volumetric Images

In this work, we introduce a method named creeping contours for image segmentation into component parts for the purpose of extracting the boundary surfaces of these parts. Creeping contours are contours that expand following a speed function defined by the gradient and curvature at contour points, starting from an initial contour position defined either manually or automatically. Contours in the image creep simultaneously at different speeds, while labels are assigned to contour pixels by the defined creeping condition. We also demonstrate the effectiveness of the proposed method by segmenting 2D grayscale images and 3D volumetric computed tomography images of mechanical parts into multiple segments and generating the boundary surfaces of these parts.

Novel Heidelberg Retina Tomograph–Based Morphological Parameters Derived from Optic Disc Cupping Surface Processing

To explore new features of the optic nerve head morphology using the Heidelberg retina tomograph (HRT) and to assess their discriminating power between glaucomatous patients and normal subjects. HRT reports, exported as TIFF images, from 97 normal subjects and 97 primary open-angle glaucoma (POAG) patients were used. For each image the contour of the dominant region of the optic disc cupping surface (dROCS) was transformed into a data series by calculating the distance of each contour pixel from the centroid. The length of contour (LC) and SD of contour (SDC) along with the dROCS area divided by the disc area (DA) HRT parameter were examined as novel parameters. The means of LC and SDC, after adjustment for cup area (CA) and DA HRT parameters, and dROCS/DA, after adjustment for CA, presented statistically significant differences (ANCOVA, P < 0.001) between the two groups. Using LC and SDC together in discriminant analysis with leave-one-out cross-validation, 75.3% of cases were correctly classified. Using dROCS/DA together with SDC, the correct classification percentage was 80.6%. The area under the ROC curve was 0.782 for LC, 0.725 for SDC, 0.861 for dROCS/DA, and 0.879 for the linear discrimination function that combines dROCS/DA and SDC. These findings suggest that LC, SDC, and dROCS/DA can be exploited to the discrimination between glaucomatous and normal subjects. LC and SDC seem to arise from the difference in the shape of the contour of dROCS between the groups, suggesting bigger deviations and irregularities in the POAG group.

고정 카메라에서의 시공간적 경계 정보를 이용한 이동 객체 윤곽선 검출 방법

In this paper, we propose a new method for detection moving object contour using spatial and temporal edge. In general, contour pixels of the moving object are likely present around pixels with high gradient value along the time axis and the spatial axis. Therefore, we can detect the contour of the moving objects by finding pixels which have high gradient value in the time axis and spatial axis. In this paper, we introduce a new computation method, termed as temporal edge, to compute an gradient value along the time axis for any pixel on an image. The temporal edge can be computed using two input gray images at time  and    using the Sobel operator. Temporal edge is utilized to detect a candidate region of the moving object contour and then the detected candidate region is used to extract spatial edge information. The final contour of the moving object is detected using the combination of these two edge information, which are temporal edge and spatial edge, and then the post processing such as a morphological operation and a background edge removing procedure are applied to remove noise regions. The complexity of the proposed method is very low because it dose not use any background scene and high complex operation, therefore it can be applied to real-time applications. Experimental results show that the proposed method outperforms the conventional contour extraction methods in term of processing effort and a ghost effect which is occurred in the case of entropy method.

Multi-oriented Bangla and Devnagari text recognition

There are printed complex documents where text lines of a single page may have different orientations or the text lines may be curved in shape. As a result, it is difficult to detect the skew of such documents and hence character segmentation and recognition of such documents are a complex task. In this paper, using background and foreground information we propose a novel scheme towards the recognition of Indian complex documents of Bangla and Devnagari script. In Bangla and Devnagari documents usually characters in a word touch and they form cavity regions. To take care of these cavity regions, background information of such documents is used. Convex hull and water reservoir principle have been applied for this purpose. Here, at first, the characters are segmented from the documents using the background information of the text. Next, individual characters are recognized using rotation invariant features obtained from the foreground part of the characters. For character segmentation, at first, writing mode of a touching component (word) is detected using water reservoir principle based features. Next, depending on writing mode and the reservoir base-region of the touching component, a set of candidate envelope points is then selected from the contour points of the component. Based on these candidate points, the touching component is finally segmented into individual characters. For recognition of multi-sized/multi-oriented characters the features are computed from different angular information obtained from the external and internal contour pixels of the characters. These angular information are computed in such a way that they do not depend on the size and rotation of the characters. Circular and convex hull rings have been used to divide a character into smaller zones to get zone-wise features for higher recognition results. We combine circular and convex hull features to improve the results and these features are fed to support vector machines (SVM) for recognition. From our experiment we obtained recognition results of 99.18% (98.86%) accuracy when tested on 7515 (7874) Devnagari (Bangla) characters.

Automated segmentation of optic disc region on retinal fundus photographs: Comparison of contour modeling and pixel classification methods

The automatic determination of the optic disc area in retinal fundus images can be useful for calculation of the cup-to-disc (CD) ratio in the glaucoma screening. We compared three different methods that employed active contour model (ACM), fuzzy c-mean (FCM) clustering, and artificial neural network (ANN) for the segmentation of the optic disc regions. The results of these methods were evaluated using new databases that included the images captured by different camera systems. The average measures of overlap between the disc regions determined by an ophthalmologist and by using the ACM (0.88 and 0.87 for two test datasets) and ANN (0.88 and 0.89) methods were slightly higher than that by using FCM (0.86 and 0.86) method. These results on the unknown datasets were comparable with those of the resubstitution test; this indicates the generalizability of these methods. The differences in the vertical diameters, which are often used for CD ratio calculation, determined by the proposed methods and based on the ophthalmologist's outlines were even smaller than those in the case of the measure of overlap. The proposed methods can be useful for automatic determination of CD ratios.

A system for the automatic estimation of morphometric parameters of corneal endothelium in alizarine red-stained images

Background/aimsA computer program for the automatic estimation of endothelium morphometric parameters (cell density, pleomorphism, polymegethism) in alizarine red-stained images is presented and evaluated.MethodsImages of corneal endothelium from 30 porcine eyes...

Diaphragm motion quantification in megavoltage cone‐beam CT projection images

To quantify diaphragm motion in megavoltage (MV) cone-beam computed tomography (CBCT) projections. User identified ipsilateral hemidiaphragm apex (IHDA) positions in two full exhale and inhale frames were used to create bounding rectangles in all other frames of a CBCT scan. The bounding rectangle was enlarged to create a region of interest (ROI). ROI pixels were associated with a cost function: The product of image gradients and a gradient direction matching function for an ideal hemidiaphragm determined from 40 training sets. A dynamic Hough transform (DHT) models a hemidiaphragm as a contour made of two parabola segments with a common vertex (the IHDA). The images within the ROIs are transformed into Hough space where a contour's Hough value is the sum of the cost function over all contour pixels. Dynamic programming finds the optimal trajectory of the common vertex in Hough space subject to motion constraints between frames, and an active contour model further refines the result. Interpolated ray tracing converts the positions to room coordinates. Root-mean-square (RMS) distances between these positions and those resulting from an expert's identification of the IHDA were determined for 21 Siemens MV CBCT scans. Computation time on a 2.66 GHz CPU was 30 s. The average craniocaudal RMS error was 1.38 +/- 0.67 mm. While much larger errors occurred in a few near-sagittal frames on one patient's scans, adjustments to algorithm constraints corrected them. The DHT based algorithm can compute IHDA trajectories immediately prior to radiation therapy on a daily basis using localization MVCBCT projection data. This has potential for calibrating external motion surrogates against diaphragm motion.

Identification of the optic nerve head with genetic algorithms

This work proposes creating an automatic system to locate and segment the optic nerve head (ONH) in eye fundus photographic images using genetic algorithms. Domain knowledge is used to create a set of heuristics that guide the various steps involved in the process. Initially, using an eye fundus colour image as input, a set of hypothesis points was obtained that exhibited geometric properties and intensity levels similar to the ONH contour pixels. Next, a genetic algorithm was used to find an ellipse containing the maximum number of hypothesis points in an offset of its perimeter, considering some constraints. The ellipse thus obtained is the approximation to the ONH. The segmentation method is tested in a sample of 110 eye fundus images, belonging to 55 patients with glaucoma (23.1%) and eye hypertension (76.9%) and random selected from an eye fundus image base belonging to the Ophthalmology Service at Miguel Servet Hospital, Saragossa (Spain). The results obtained are competitive with those in the literature. The method's generalization capability is reinforced when it is applied to a different image base from the one used in our study and a discrepancy curve is obtained very similar to the one obtained in our image base. In addition, the robustness of the method proposed can be seen in the high percentage of images obtained with a discrepancy delta<5 (96% and 99% in our and a different image base, respectively). The results also confirm the hypothesis that the ONH contour can be properly approached with a non-deformable ellipse. Another important aspect of the method is that it directly provides the parameters characterising the shape of the papilla: lengths of its major and minor axes, its centre of location and its orientation with regard to the horizontal position.

Triangle refinement in a constrained Delaunay triangulation skeleton

This paper presents an algorithm with the purpose of improving upon the already successful constrained Delaunay triangulation (CDT) skeletonisation technique. Using such a triangulation to construct a skeleton has proven very effective, that can sometimes, however, produce triangles that do not represent the true nature of the underlying shape. The contour pixels chosen for triangulation are of significant importance, as they determine the triangle edges that define the skeleton. The algorithm described in this paper deals with this problem by inserting new triangulation points in strategic locations in end, normal and junction triangles. Results show that the skeletons produced by this algorithm are accurate, robust against noise and, above all, comply much better with a human's perception of the image than the original triangulation method.

Efficient skeletonization based on generalized discrete local symmetries

This paper presents an efficient skeletonization method based on generalized discrete local symmetries. A generalized discrete local symmetry is a local symmetry between a contour pixel and a contour segment on the opposite side of the underlying shape. The centerlines of the local symmetries of a shape form the skeleton of the shape. The proposed method is fast compared to two existing skeletonization methods. A speed-up factor of more than 50 can be achieved for high-resolution images. The method is also robust against noise and geometrical transformations, such as rotation and uniform scaling. The method is suitable for skeletonizing high-resolution images where it can be impractical to use other skeletonization techniques, such as thinning and distance transforms, because of the high computational complexity.

Skeletonization based on error reduction

The accuracy of a non-pixel-based skeletonization method is largely dependent on the contour information chosen as input. When using a Constrained Delaunay Triangulation to construct an object's skeleton, a number of contour pixels must be chosen as a basis for triangulation. This paper presents a new method of selecting these contour pixels. A new method for measuring skeletonization error is proposed, which quantifies the deviation of a skeleton segment from the true medial axis of a stroke in an image. The goal of the proposed algorithm is to reduce this error to an acceptable level, whilst retaining the superior efficiencies of previous non-pixel-based techniques. Experimental results show that the proposed method is adept at following the medial axis of an image, and is capable of producing a skeleton that is confirmed by a human's perception of the image. It is also computationally efficient and robust against noise.

Sci‐AM1 Sat ‐ 05: Fractal and motion modeling of PET/CT tumours

Objective: To generate realistic computer‐simulated PET/CT tumour images with characteristics typical of the tumour (shape, pixel intensity and movement over time) and of the imaging system (acquisition parameters and resolution). Methods: An initial tumour volume was segmented from the CT scan of a lung cancer patient via region growing. To improve the axial resolution and mimic the contour curvature and pixel variations in the transaxial plane, fractal analysis and interpolation was used. Breathing motion with a period and extents typical of lung cancer patients was then applied to the fractal tumour. To produce PET and CT images, the moving tumour was sampled according to acquisition characteristics typical of the two modalities. For CT, a slice thickness of 3.0mm, acquisition time of 0.7s and resolution of 2mm was assumed. For PET, it was assumed that tumour motion was captured within one bed position over an acquisition time of 180s, and the imager resolution was 6mm. Results: Following fractal interpolation, the tumour exhibited similar curvature and gray‐level gradation in all three imaging planes. Application of motion and blurring produced CT images different in appearance from the stationary fractal tumour volume and the total volume traced by the moving object. Motion and blurring produced PET images similar to the volume traced by the moving object, graded by the time spent in each location within the volume and blurred due to the degraded resolution of the scanner. Conclusions: Realistic computer‐simulated PET/CT tumour images can be generated using fractal analysis and motion modeling.

Robust Face Image Matching under Illumination Variations

Face image matching is an essential step for face recognition and face verification. It is difficult to achieve robust face matching under various image acquisition conditions. In this paper, a novel face image matching algorithm robust against illumination variations is proposed. The proposed image matching algorithm is motivated by the characteristics of high image gradient along the face contours. We define a new consistency measure as the inner product between two normalized gradient vectors at the corresponding locations in two images. The normalized gradient is obtained by dividing the computed gradient vector by the corresponding locally maximal gradient magnitude. Then we compute the average consistency measures for all pairs of the corresponding face contour pixels to be the robust matching measure between two face images. To alleviate the problem due to shadow and intensity saturation, we introduce an intensity weighting function for each individual consistency measure to form a weighted average of the consistency measure. This robust consistency measure is further extended to integrate multiple face images of the same person captured under different illumination conditions, thus making our robust face matching algorithm. Experimental results of applying the proposed face image matching algorithm on some well-known face datasets are given in comparison with some existing face recognition methods. The results show that the proposed algorithm consistently outperforms other methods and achieves higher than 93% recognition rate with three reference images for different datasets under different lighting conditions.

Perceptually stable regions for arbitrary polygons

Zou and Yan have recently developed a skeletonization algorithm of digital shapes based on a regularity/singularity analysis; they use the polygon whose vertices are the boundary pixels of the image to compute a constrained Delaunay triangulation (CDT) in order to find local symmetries and stable regions. Their method has produced good results but it is slow since its complexity depends on the number of contour pixels. This paper presents an extension of their technique to handle arbitrary polygons, not only polygons of short edges. Consequently, not only can we achieve results as good as theirs for digital images, but we can also compute skeletons of polygons of any number of edges. Since we can handle polygonal approximations of figures, the skeletons are more resilient to noise and faster to process.

Research on scale effect of histogram

To describe the spatial relationship among the earth objects compactly, in this paper, we raised the concept of histo-variogram based on the analysis of the characteristics of other spatial analyzing methods such as variogram, information entropy. And we also raised a new spatial analysis method of histogram decomposition based on the definition of standing pixel and contour pixel. At the end of this paper, we demonstrated the characteristics of histo-variogram by two experiments, one for spatial analysis, the other for image fusion.

Corner detection by means of contour local vectors

A corner detection algorithm for planar shapes based on a new curvature index is proposed. The calculation parameters of this new index are adapted to the shape curvature at each point of its contour. The process consists of estimating the maximum length of contour yielding no significant discontinuities on the right and left sides of each contour point in order to calculate the contour curvature index at that point. Thus, the curvature index is adaptively filtered depending on the natural scale of the contour pixels. Corners are detected by simply thresholding the curve. The proposed method provides a more precise characterisation of the contour. Detected corners are very stable against noise distortion, scaling or rotation.

The rate of phosphocreatine hydrolysis and resynthesis in exercising muscle in humans using 31P-MRS.

Time-resolved 31-phosphorus nuclear magnetic resonance spectroscopy (31P-MRS) of the biceps femoris muscles was performed during exercise and recovery in six healthy sedentary male subjects (maximal oxygen uptake; 46.6 +/- 1.7 (SEM) ml.kg-1.min-1), 5 male sprinters (56.2 +/- 2.5), and 5 male long-distance runners (73.6 +/- 2.2). Each performed 4 min of knee flexion exercises at absolute values of 1.63 W and 4.90 W, followed by 5 min of recovery in a prone position in a 2.1 T superconducting magnet with a 67 cm bore. 31P-MRS spectra were recorded every 12.8 s during the rest-exercise-recovery sequence. Computer-aided contour analysis and pixel imaging of phosphocreatine peaks (PCr) and inorganic phosphate (Pi) were performed. The work loads in the present study were selected as mild exercise (1.63 W) and heavy exercise (4.90 W), corresponding to 18-23% and 54-70% of maximal exercise intensity. Long-distance runners showed a significantly smaller decrement in PCr and less acidification at a given exercise intensity compared to those shown by sedentary subjects. The transient responses of PCr and Pi during recovery were characterized by first-order kinetics. After exercise, the recovery rates of PCr and Pi were significantly faster in long-distance runners than in sedentary subjects (P < 0.05). Since it is postulated that PCr resynthesis is controlled by aerobic metabolism and mitochondrial creatine kinase, it is suggested that the faster PCr and Pi recovery rates and decreased acidification seen in long-distance runners during and after exercise might be attributed to their greater capacity for aerobic metabolism.

Application of 3D computer-assisted techniques to sinonasal pathology—Case report: War wounds of paranasal sinuses caused by metallic foreign bodies

Foreign bodies in paranasal sinuses are found quite infrequently. They are usually detected after various types of head trauma and most commonly occur as a consequence of improper handling of firearms or explosives. In countries at war, eg, during the war in Croatia, adults and children were almost equally exposed to these injuries. The diagnosis should be based on results from precise computer tomography (CT) scanning in axial and coronal sections, and, using these data, tissues of different densities at these anatomical locations can be differentiated. The possibility of exact preoperative, noninvasive visualization of the spatial relationships of anatomic and pathologic structures with 3-dimensional (3D) computer-assisted diagnosis and intraoperative navigational techniques allows the surgeon to achieve a considerable advantage in the preoperative examination of the patient and to reduce the risk of intraoperative complications, all by the use of virtual surgery (VS) or virtual diagnosis. The expected contribution of the mentioned computer-assisted surgical technique manifests itself in defining the most appropriate mode of CT scanning of the head to design the 3D operating field model, and the possibility of active and dynamic 3D visualization of the desired anatomical regions is realized. 3D reconstruction of anatomic units becomes a routine preoperative procedure, providing a highly useful and informative visualization of the regions of interest, and, thus, advancing the definition of geometric information on anatomical contours of the 3D model by the transfer of so-called image pixel to contour pixel. (Am J Otolaryngol 2002;23:27-34. Copyright © 2002 by W.B. Saunders Company)

Optimized approach to multiscale skeleton generation

We describe a simple, accurate, and fast approach to obtaining high-quality 2-D multiscale symmetry axis transform (SAT) skeletons based on the exact Euclidean distance. The method involves labeling the contour pixels of the shape and propagating such labels only around the neighborhood of already obtained skeleton segments. Since only small neighborhoods are processed, the algorithm efficiency becomes maximized.