Landslide Susceptibility Mapping through Weightages Derived from Statistical Information Value Model
An Efficient Foot Ulcer Determination System for Diabetic Patients
Statistical Wavelet based Adaptive Noise Filtering Technique for MRI Modality
Real Time Sign Language: A Review
Remote Sensing Schemes Mingled with Information and Communication Technologies (ICTS) for Flood Disaster Management
FPGA Implementation of Shearlet Transform Based Invisible Image Watermarking Algorithm
A Comprehensive Study on Different Pattern Recognition Techniques
User Authentication and Identification Using NeuralNetwork
Flexible Generalized Mixture Model Cluster Analysis withElliptically-Contoured Distributions
Efficient Detection of Suspected areas in Mammographic Breast Cancer Images
Breast cancer is the most common type of cancers found in women all across the world. Mammography is considered as an effective tool for early detection and diagnosis of breast cancer. The most common breast abnormalities that may indicate breast cancer are masses and micro-calcifications or calcifications. The abnormalities present in breast images are characterized by using a range of features that may be missed or misinterpreted by radiologists while reading large amount of mammographic images during cancer screening process. Computer-aided diagnosis (CAD) systems have been developed to assist radiologists to provide an accurate diagnosis. An attempt has been made to improve the classification performance of CAD system in which shape and texture are used in analyzing region of interest (ROI) of mammographic images of breast. The method detects ROI by combining edge and region criteria and then feature extraction method helps extract few statistical parameters such as sensitivity and specificity to evaluate the performance of the proposed method. The sensitivity of the proposed method is 97.5% and specificity is 91.2% that produced an accuracy of 96.6%. Size of tumor is also computed and classification stage of breast cancer is identified.
The Ridgelet Transform [6] was introduced as a sparse expansion for functions of continuous spaces that are smooth away from discontinuities along lines. The powerful properties of the ridgelets are catching and representing monodimensional singularities in bi-dimensional space [8]. Using these effective properties, in this paper the authors propose an effect of Ridgelet Transform on various Similarity/Distance Measure Techniques namely Euclidean Distance, Modified Squared Euclidean Distance, Correlation Distance and Angle Distance for an unconstrained bi-lingual handwritten character recognition. Ridgelet Transform is used to extract a character image of low pass energy and is then fed to PCA for feature extraction. We conducted experiment on very large database of bi-lingual handwritten characters (Kannada and English). The database contains the samples of 22,600 and the effect of the proposed method is compared with the standard PCA & FLD methods. Among the above mentioned similarity/distance measure techniques the better recognition accuracy were achieved using angle distance measure.
In this paper, Conditional Random Field Based Segmentation and different model-based Markov Random Field(MRF) classification for skin lesions in dermoscopic images are proposed. This method is used in the pattern analysis framework for diagnosis of melanoma by dermatologists. A Dermoscopic image is smoothened by Wiener Filer Method and converted into Grayscale Image. Then the image is diluted which gives the contour of an image. The input image is segmented by Conditional Random Field Technique. The Estimated CPU time is calculated which gives less Processing Time. Then classification is carried out by an image retrieval approach with different distance metrics. These features are supposed to follow Gaussian Model, Gaussian Mixture Model, and Bag-of-features Histogram Model. The main aim of this paper is the classification of an entire pigmented lesion and analysis of the texture of an image. The image database is extracted from a public Atlas of Dermoscopy. Receiver Operating Characteristics (ROC) Curve is used to evaluate the performance of Segmentation Process which gives more accuracy. Finally, the skin lesions with their levels were analysed.
There are several application areas where tremendous computational resources are required including image processing, big data and genetic mapping which are computationally intensive areas. Huge computing resources are required to solve such complex problems and powerful computing environment is needed. An emphasis is made on fractal image compression, which requires higher computing needs to solve. Single Instruction Multiple Data approach is followed using distributed architecture. The research compares the performance on the basis of speed up and encoding time. It was found that image compression requires more computing power to solve in lesser time. In this paper, the parallel algorithms are developed using Distributed Fractal Image Encoding Architecture (DFIE) as Single Instruction Multiple Data (SIMD) approach
Efficacious Recognition and consistent identification of visual features is an important problem in applications, such as Object Recognition, Structure from Motion, Image Indexing and Visual Localization. The input data takes, many forms such as Video Sequences, views from Multiple Cameras or Multi-dimensional data from a scanner. Concurrent performance is a perilous demand to utmost of these applications, which necessitate the finding and corresponding of the visual features in real time. Although Feature Recognition and Identification Methods have been studied in the Literature, due to their Computational Intricacy, pure software execution by Unique Hardware is far from suitable in their performance for Real Time Applications. The existing system consists of Scale Invariant Feature Transform (SIFT) for Feature Detection and Binary Robust Independent Elementary Features (BRIEF) for Feature Description and Matching. This system fails to detect the features which are invariant to scale, change in viewpoint and illumination, and the addition of noise. The proposed system consists of Wavelet Feature Extraction Method, and for classification process, Subtractive Clustering is used. This system reduces time consumption and overall system complexity. This paper focuses on a different hardware design to enable real-time performance of founding correspondences between idealistic sequential frames of high-resolution 720 p (1280 x 720) video. Due to these assistances, the proposed system attains feature detection and matching at 60 frame/s for 720-p video. Its processing speed can encounter and even overdo the demand of most realistic concurrent video analytics applications.
