International Journal of Image and Graphics

Fiber directional tracking through diffusion tensor magnetic resonance imaging (DT-MRI) is a promising research field in visualization and computer graphics and is widely applied in the reconstruction of fiber orientation and the structure of biological tissues. The filter technique used to blur noise in data is of critical importance for fiber directional tracking, particularly because existing tracking methods are very sensitive to impulsive noise. In this paper, a mixed filter of the 3D Gauss and directional distance filter (GDDF) is proposed to suppress noises in corrupted vector fields. Simulation results and objective evaluation of vector datasets demonstrate that GDDF not only possesses the capability of noise attenuation but also preserves vector directions. By validating the simulated vector data against experimental heart data, it is also shown that the GDDF is an effective and stable preprocessing method that accurately reconstructs fiber orientation.

Recorded medical images often represent a degraded version of the original scene due to imperfections in electronic or photographic medium used. The degradations may have many causes, but two dominant degradations are noise and blur. Restoration of blurred and noisy medical images is of fundamental importance in several medical imaging applications. Most of the medical image denoising techniques need removal of blur before the denoising. Denoising of medical images in presence of blur is a hard problem. Most of the wavelet transform-based denoising techniques use the orthonormal wavelets and suitable for image corrupted with only additive white Gaussian noise. In the present work, we have proposed a denoising algorithm for medical images based on the lifting-scheme and linear phase characteristic of biorthogonal wavelet transform. A level-dependent soft thresholding function has been used which is based on the standard deviation, the absolute mean and the absolute median of the wavelet coefficients. The linear phase characteristic of the biorthogonal filters used in denoising reduces the distortions at edge points of image. Also, the lifting schemes of the biorthogonal wavelet filters make the algorithm efficient and applicable in real time. Experimental results show that the proposed denoising method outperform standard wavelet, complex wavelet and curvelet-based denoising techniques in terms of the SNR and PSNR (in dB) and it offers effective noise removal from noisy medical images while maintaining sharpness of objects in the image.

Reliable extraction/segmentation of text lines, words and characters is one of the very important steps for development of automated systems for understanding the text in low resolution display board images. In this paper, a new approach for segmentation of text lines, words and characters from Kannada text in low resolution display board images is presented. The proposed method uses projection profile features and on pixel distribution statistics for segmentation of text lines. The method also detects text lines containing consonant modifiers and merges them with corresponding text lines, and efficiently separates overlapped text lines as well. The character extraction process computes character boundaries using vertical profile features for extracting character images from every text line. Further, the word segmentation process uses k-means clustering to group inter character gaps into character and word cluster spaces, which are used to compute thresholds for extracting words. The method also takes care of variations in character and word gaps. The proposed methodology is evaluated on a data set of 1008 low resolution images of display boards containing Kannada text captured from 2 mega pixel cameras on mobile phones at various sizes 240 × 320, 480 × 640 and 960 × 1280. The method achieves text line segmentation accuracy of 97.17%, word segmentation accuracy of 97.54% and character extraction accuracy of 99.09%. The proposed method is tolerant to font variability, spacing variations between characters and words, absence of free segmentation path due to consonant and vowel modifiers, noise and other degradations. The experimentation with images containing overlapped text lines has given promising results.

Ray tracing of implicit surfaces based on radial basis functions can demand high computational cost in the presence of a large number of radial centers. Recently, it was presented the least squares hermite radial basis functions (LS-HRBF) Implicits, a method for implicit surface reconstruction from Hermitian data (points equipped with their normal vectors) which makes use of iterative center selection in order to reduce the number of centers. In the present work, we propose an antialiazed sphere tracing algorithm fully implemented in OpenGL Shader Language for ray tracing LS-HRBF Implicits, which exploits a regular partition of unity for strong parallelization. We show that interactive frame rates can be achieved for surfaces composed of thousands of centers even when rendering effects such as cube mapping, soft shadows and ambient occlusion are used.

Image segmentation is a fundamental step in several image processing tasks. It is a process where an image is divided into its constituent regions guided by a similarity criterion. One very interesting image segmentation method is the color structure code (CSC), which combines simultaneously split-and-merge and region-growing techniques. In this paper, a segmentation approach based on the CSC method, weighted color structure code (WCSC), is proposed. This method is guided by a nonlinear discrimination function, where the user-inference is captured by the Polynomial Mahalanobis distance, prioritizing, during the merging process, the regions with higher similarity to the user selected pattern. The WCSC has color distribution pattern-oriented characteristic, showing better coherence among the segments with higher similarity to the selected pattern. A qualitative evaluation and parametric paired analysis were performed to compare CSC, WCSC and other segmentation methods results, using images from Berkeley benchmark. The results from these comparison indicate an improvement on the segmentation result obtained by the WCSC.

This paper presents a novel method for computer vision-based static and dynamic hand gesture recognition. Haar-like feature-based cascaded classifier is used for hand area segmentation. Static hand gestures are recognized using linear discriminant analysis (LDA) and local binary pattern (LBP)-based feature extraction methods. Static hand gestures are classified using nearest neighbor (NN) algorithm. Dynamic hand gestures are recognized using the novel text-based principal directional features (PDFs), which are generated from the segmented image sequences. Longest common subsequence (LCS) algorithm is used to classify the dynamic gestures. For testing, the Chinese numeral gesture dataset containing static hand poses and directional gesture dataset containing complex dynamic gestures are prepared. The mean accuracy of LDA-based static hand gesture recognition on the Chinese numeral gesture dataset is 92.42%. The mean accuracy of LBP-based static hand gesture recognition on the Chinese numeral gesture dataset is 87.23%. The mean accuracy of the novel dynamic hand gesture recognition method using PDF on directional gesture dataset is 94%.

Efficient interactive foreground extraction from color images is of great practical importance in computer vision. In recent years, an approach based on optimization using graph-cuts has been widely used. However, an interactive foreground extraction approach, which is intuitive and efficient for users, is required for practical applications. In this paper, we propose a novel foreground object extraction method based on interactive color saliency map. In our method, the user provides only some foreground pixels as the initial reference. To achieve extraction of multi target objects, likelihoods of the foreground and background are defined by a color saliency map based on the provided reference pixels. These likelihoods are added to the cost of the proposed graph. Finally, image segmentation is performed by optimizing the proposed graph using the graph-cut algorithm.