Rajiv Kapoor

Image enhancement using Exposure based Sub Image Histogram Equalization

This paper presents a novel Exposure based Sub-Image Histogram Equalization (ESIHE) method for contrast enhancement for low exposure gray scale image. Exposure thresholds are computed to divide the original image into sub-images of different intensity levels. The histogram is also clipped using a threshold value as an average number of gray level occurrences to control enhancement rate. The individual histogram of sub images is equalized independently and finally all sub images are integrated into one complete image for analysis. The simulation results show that ESIHE outperforms other conventional Histogram Equalization (HE) methods in terms of image visual quality, entropy preservation and better contrast enhancement.

A new algorithm for skew detection and correction

Here we have proposed two algorithms. The first one detects the skewing of words and the second corrects the skewing from handwritten words. Both algorithms make use of the Radon transform based projection profile technique. The method does not require pre-processing and it works equally good even with noise. The method is fast. The algorithms have been tested on words taken from more than 200 writers and the results obtained confirm the overall accuracy of proposed system. No error was detected.

Intelligent video target tracking using an evolutionary particle filter based upon improved cuckoo search

Abstract The aim of this paper is to propose an evolutionary particle filter based upon improved cuckoo search algorithm which will overcome the sample impoverishment problem of generic particle filter. In our proposed method, improved cuckoo search (ICS) algorithm is embedded into particle filter (PF) framework. Improved cuckoo search algorithm uses levy flight for generating new particles in the solution and introduced randomness in samples by abandoning a fraction of these particles. The second important contribution in this article is introduction of new way for tackling scaling and rotational error in object tracking. Performance of proposed improved cuckoo particle filter is investigated and evaluated on synthetic and standard video sequences and compared with the generic particle filter and particle swarm optimization based particle filter. We show that object tracking using improved cuckoo particle filter provides more reliable and efficient tracking results than generic particle filter and PSO-particle filter. The proposed technique works for real time video objects tracking.

Recent advances on multicue object tracking: a survey

The performance of single cue object tracking algorithms may degrade due to complex nature of visual world and environment challenges. In recent past, multicue object tracking methods using single or multiple sensors such as vision, thermal, infrared, laser, radar, audio, and RFID are explored to a great extent. It was acknowledged that combining multiple orthogonal cues enhance tracking performance over single cue methods. The aim of this paper is to categorize multicue tracking methods into single-modal and multi-modal and to list out new trends in this field via investigation of representative work. The categorized works are also tabulated in order to give detailed overview of latest advancement. The person tracking datasets are analyzed and their statistical parameters are tabulated. The tracking performance measures are also categorized depending upon availability of ground truth data. Our review gauges the gap between reported work and future demands for object tracking.

Contrast enhancement via texture region based histogram equalization

Abstract This paper presents two novel contrast enhancement approaches using texture regions-based histogram equalization (HE). In HE-based contrast enhancement methods, the enhanced image often contains undesirable artefacts because an excessive number of pixels in the non-textured areas heavily bias the histogram. The novel idea presented in this paper is to suppress the impact of pixels in non-textured areas and to exploit texture features for the computation of histogram in the process of HE. The first algorithm named as Dominant Orientation-based Texture Histogram Equalization (DOTHE), constructs the histogram of the image using only those image patches having dominant orientation. DOTHE categories image patches into smooth, dominant or non-dominant orientation patches by using the image variance and singular value decomposition algorithm and utilizes only dominant orientation patches in the process of HE. The second method termed as Edge-based Texture Histogram Equalization, calculates significant edges in the image and constructs the histogram using the grey levels present in the neighbourhood of edges. The cumulative density function of the histogram formed from texture features is mapped on the entire dynamic range of the input image to produce the contrast-enhanced image. Subjective as well as objective performance assessment of proposed methods is conducted and compared with other existing HE methods. The performance assessment in terms of visual quality, contrast improvement index, entropy and measure of enhancement reveals that the proposed methods outperform the existing HE methods.

Fusion for registration of medical images - a study

The paper is a study demonstrating the application of discrete multiwavelets in medical image registration. The idea is to improve the image content by fusing images like MRI, CT and SPECT images, so as to provide more information to the doctor. The process of fusion is not new but here the results of study have been compared with the results from FCM algorithm used for similar application. Multiwavelets have been used for better clustering, as their decomposition results were better than Daubechies decomposition. A new feature based fusion algorithm has been used. This method shows results better than other methods for image registration when the images have been taken for the same person at a particular angle. The selective fusion not only gives more information but also helps in disease detection.

A proposed unified framework for the recognition of human activity by exploiting the characteristics of action dynamics

The aim of this paper is to present a novel integrated framework for the recognition of human actions using a spatial distribution of edge gradient (SDEG) of human pose and detailed geometric orientation of a human silhouette in a video sequence. The combined descriptor endows a wealthy feature vector dictionary having both the appearance and angular kinematics information that significantly wraps the local and global information and provides discriminative depiction for the action recognition. The SDEG is computed on a still image at different levels of resolution of sub-images, and still images of the human poses are extracted from the input video sequence using fuzzy trapezoidal membership function based on the normalized histogram distance between the contiguous segment frames. The change of geometric orientation of human silhouette with time is computed using normalized R -Transform. To validate the performance of the proposed approach, extensive experiments are conducted on five publicly available human action datasets i.e. Weizmann, KTH, Ballet Movements, Multi-view i3dPost, and IXMAS. The recognition accuracy achieved on these datasets demonstrates that the proposed approach has an abundant discriminating power of recognizing the variety of actions. Moreover, the proposed approach yields superior results when compared with similar state-of-the-art methods. A combined algorithm based on shape and motion features of human activity.A single key pose is used for estimation of shape using edges.A single global key pose is extracted from video signal by exploiting local notion.The temporal motion feature is computed using R -transform.Robustness of the algorithm is demonstrated on the varied dataset.

Fingerprint denoising using ridge orientation based clustered dictionaries

In this paper, we introduce a novel approach of fingerprint denoising using ridge orientation based clustered sub dictionaries. The idea behind this approach is to group the patches of similar geometric structures or dominant orientation and construct separate sub-dictionaries for each group. The orientation of ridge or a valley has been exploited in finger print matching algorithms in the past. In the proposed method, the same concept of ridge orientation is utilized to group the patches and to subdivide a large dictionary into array of sub dictionaries. The new approach undergoes three steps i.e. ridge orientation based clustering, dictionary learning and sparse coefficient calculation. While reconstructing the image in final step, the minimum residual error criterion is used for choosing sub dictionary for a particular patch. The algorithm performance is experimentally compared with other existing methods not only in terms of PSNR, SSIM measures but also in terms of Euclidean distance parameter, which is used, for fingerprint matching procedures. The simulation results demonstrates that the new method achieves better results in comparison with its counterparts and will establish in improving performances of fingerprint-identification systems. Dominant orientation based sub dictionaries for effectiveness of sparse modeling.Sub dictionary for smooth patches (no dominant orientation) improves reconstruction.Minimum residue error criterion based selection of sub dictionary for reconstruction.

Fingerprint image super-resolution via ridge orientation-based clustered coupled sparse dictionaries

Abstract. The process of image quality improvement through super-resolution methods is still a gray area in the field of biometric identification. This paper proposes a scheme for fingerprint super-resolution using ridge orientation-based clustered coupled sparse dictionaries. The training image patches are clustered into groups based on dominant orientation and corresponding coupled subdictionaries are learned for each low- and high-resolution patch groups. While reconstructing the image, the minimum residue error criterion is used for choosing a subdictionary for a particular patch. In the final step, back projection is applied to eliminate the discrepancy in the estimate due to noise or inaccuracy in sparse representation. The performance evaluation of the proposed method is accomplished in terms of peak signal-to-noise ratio and structural similarity index. A filter bank-based fingerprint matcher is used for evaluating the performance of the proposed method in terms of matching accuracy. Our experimental results show that the new method achieves better results in comparison with other methods and will establish itself for improving performances of fingerprint-identification systems.

Integrated approach for human action recognition using edge spatial distribution, direction pixel and -transform

In this article, a simple yet proficient approach for the recognition of human action and Activity is presented. This method is based on the integration of translation and rotation of the human body. The proposed framework undergoes three major steps: (i) the shape of the human action/activity is represented through the computation of average energy images using edge spatial distribution of gradients along with the directional variation of the pixel values, (ii) the orientation-based rotational information of the human action is computed through -transform and (iii) a descriptor is developed by the fusion of translational features with rotational features. The fusion of features possesses the advantages exhibited by both local and global features of the silhouette and thus provides the discriminating feature representation for human activity recognition. The performance of descriptor is evaluated through a hybrid approach of support vector machine and the nearest neighbour classifiers on standard data set. The proposed method has shown superior results in terms of recognition accuracy in comparison with other state-of-the-art methods.