Debabrata Samanta

Development of Edge Detection Technique for Images Using Adaptive Thresholding

Edge detection is a terminology in image processing and computer vision, particularly in the areas of feature detection and feature extraction, to refer to algorithms which aim at identifying points in a digital image at which the image brightness changes sharply or more formally has discontinuities. In this paper a simple edge detection technique is performed by convoluting the image using a five order mask and then Selecting a threshold from taking the sum of mean value and the standard deviation of the gradient image within a 3 X 3 window. In Canny edge detection a gauss kernel is used for smoothing the image in order to eliminate the noise. But in our algorithm a single mask perform the noise smoothing as well as finding the edge strength.

Admission control algorithm based on the effective bandwidth in vehicle-to-infrastructure communication.

With pervasive of mobile computing and wireless communication, the vehicular ad-hoc network has become a key technology in modern days information exchange. Due to the fast growing number of vehicles and saturation of the transport infrastructure, it is inhabitable to come out with better solution considering the available resources. In this study, thorough traffic analysis has been carried out to suggest a better model. Further, an attempt has been made to propose an admission control algorithm based-on the parameters evolve from the mathematical analysis. Moreover, the performance analysis has been shown to establish the efficacy of the algorithm.

Statistical Approach for Extraction of Panic Expression

Panic is an impetuous occurrence that can emerge on the human face based on any abrupt situation or atypical circumference. The facial expression of human being is nothing but resembles a reflection of the circumstances or consequences those he or she is facing or yet to face. The happening of panic has become an alarming phenomenon in the present era. Isolation of a people enriched with abnormal facial expression is one of the tremendous difficult tasks. In this paper we have proposed a new methodology based on statistical approach with respect to time span series. Firstly, we are to capture facial expressions with the help of high resolution camera and stored them in a database. After that we followed transformation for image enhancement. We have considered Wiener filter for Image Restoration. Then we are to identify the span spaces to collect the span points of the facial image with respect to nose position and we are to calculate the serial correlation for Ranking Expression. Finally, the geometric value of time span series represents the data of panic expression.

A Novel Approach of SAR Image Classification Using Color Space Clustering and Watersheds

Image Classification is the progression of separating or grouping an image into different parts. The good feat of recognition algorithms based on the quality of classified image. The good performance of recognition algorithms depend on the quality of classified image. An important problem in SAR image application is accurate classification. In this paper, we consider the problem of SAR image Classification by Histogram thresholding technique. Then we proposed Color space clustering and Watershed Classification for merging different region to get the classified SAR images.

A novel approach for panic-face extraction based on mutation

The facial expression of mankind is one of the vital evidence of the surrounded circumstances that a person is facing or overwhelmed to be faced and Panic is a spontaneous occurrence that emerges through any persons face depending on environmental state of affairs or circumferences. The Mutation can have an imperative role in the way of image recognition and its features extraction. It may carefully be measured in the structure of matrices so as to plot a space similar to Euclidean-space. The use of covariance features has better and faster results as compared to most of the methodology usually are in use on the way of object/image identification. In this paper we present a different technique for the recognition of facial expression of any persons Panic moment by Mutation. We tried to consider Mutation based on span-space of covariance matrices to replicate the belongingness of errors that could occur less, probably, during the phase of repetition. The pixels span-space-points and Mutation coefficients are engaged to outward appearance of the covariance matrices. The results set of the Experiment expose the reward of this anticipated methodology.

Novel Shannon’s Entropy Based Segmentation Technique for SAR Images

Segmentation of SAR image plays an imperative function in analysis of huge amount of satellite data. The good recital of recognition algorithms based on the quality of segmented image. In case of SAR image, it is one of the most complicated and challenging tasks in image processing, and determines the quality of the final results of the analysis. The capability of SAR image is to penetrate cloud cover to predict the weather condition at any particular instant of time. Image data can also be used to classify the land, forest, hills, oceans etc.

Extraction of Panic Expression from Human Face Based on Histogram Approach

Panic expression is one of the most important features of facial recognition in the present era, and it has become a burning issue. It is very difficult to segregate a people whether he or she is in the normal position or barring unexpected circumstances that has occurred abruptly. Over the past 50 years, different researchers have developed human-observer based methods that can be used to classify and correlate facial expressions with human sensation. In this paper we proposed a novel methodology based on histogram classification for extraction of panic moment of a human being. Firstly, we have considered face-mask to collect the maximum information from human face. Then we have set the three coordinate positions for storing the data as a panic-info-mask. Finally, the geometrical value of panic-info-mask represents the data of panic moment.

Extraction of Panic Expression Depending on Lip Detection

Expressions of a people face are the reflections of some essential evidence of his or her mission or activities. Automatically, the expressions are revealed trough different parts of face likes lip, eyes, eye-brow, forehead, etc. depending on the surrounding and environmental situation that an individual is facing or perceiving to be faced. Lip is one of the most impartial parts of Human face. Expression from a human lip is an impetuous incidence that emerges based on environmental circumferences that he or she is facing. However, in this paper we have taken lip as the parameter to extract the different types of facial expressions like happiness, fear, anger, disgust, sadness, etc. The zero crossing algorithms is used to segment the input-face-image into two regions: First one is LIP and the second one is the remaining parts of that face excluding the LIP keeping nose tip as the Base-point. The position of the segmented LIP is being tracked automatically based on Isolated System. An automated LIP detection system based on Homomorphic filter has been used with respect to Cross Correlation.