Naveed Ejaz

Adaptive key frame extraction for video summarization using an aggregation mechanism

Video summarization is a method to reduce redundancy and generate succinct representation of the video data. One of the mechanisms to generate video summaries is to extract key frames which represent the most important content of the video. In this paper, a new technique for key frame extraction is presented. The scheme uses an aggregation mechanism to combine the visual features extracted from the correlation of RGB color channels, color histogram, and moments of inertia to extract key frames from the video. An adaptive formula is then used to combine the results of the current iteration with those from the previous. The use of the adaptive formula generates a smooth output function and also reduces redundancy. The results are compared to some of the other techniques based on objective criteria. The experimental results show that the proposed technique generates summaries that are closer to the summaries created by humans.

Efficient visual attention based framework for extracting key frames from videos

The huge amount of video data on the internet requires efficient video browsing and retrieval strategies. One of the viable solutions is to provide summaries of the videos in the form of key frames. The video summarization using visual attention modeling has been used of late. In such schemes, the visually salient frames are extracted as key frames on the basis of theories of human attention modeling. The visual attention modeling schemes have proved to be effective in video summarization. However, the high computational costs incurred by these techniques limit their applicability in practical scenarios. In this context, this paper proposes an efficient visual attention model based key frame extraction method. The computational cost is reduced by using the temporal gradient based dynamic visual saliency detection instead of the traditional optical flow methods. Moreover, for static visual saliency, an effective method employing discrete cosine transform has been used. The static and dynamic visual attention measures are fused by using a non-linear weighted fusion method. The experimental results indicate that the proposed method is not only efficient, but also yields high quality video summaries.

Prioritization of brain MRI volumes using medical image perception model and tumor region segmentation

The objective of the present study is to explore prioritization methods in diagnostic imaging modalities to automatically determine the contents of medical images. In this paper, we propose an efficient prioritization of brain MRI. First, the visual perception of the radiologists is adapted to identify salient regions. Then this saliency information is used as an automatic label for accurate segmentation of brain lesion to determine the scientific value of that image. The qualitative and quantitative results prove that the rankings generated by the proposed method are closer to the rankings created by radiologists.

Feature aggregation based visual attention model for video summarization

Abstract Video summarization is an integral component of video archiving systems. It provides small versions of the videos that are suitable for enhancing browsing and navigation capabilities. A popular method to generate summaries is to extract a set of key frames from the video, which conveys the overall message of the video. This paper introduces a novel feature aggregation based visual saliency detection mechanism and its usage for extracting key frames. The saliency maps are computed based on the aggregated features and motion intensity. A non-linear weighted fusion mechanism combines the two saliency maps. On the resultant map, a Gaussian weighting scheme is used to assign more weight to the pixels close to the center of the frame. Based on the final attention value of each frame, the key frames are extracted adaptively. The experimental results, based on different evaluation standards, demonstrate that the proposed scheme extracts semantically significant key frames.

Multi-kernel based adaptive interpolation for image super-resolution

This paper proposes a cost-effective and edge-directed image super-resolution scheme. Image super-resolution (image magnification) is an enthusiastic research area and is desired in a variety of applications. The basic idea of the proposed scheme is based on the concept of multi-kernel approach. Various stencils have been defined on the basis of geometrical regularities. This set of stencils is associated with the set of kernels. The value of a re-sampling pixel is obtained by calculating the weighted average of the pixels in the selected kernel. The time complexity of the proposed scheme is as low as that of classical linear interpolation techniques, but the visual quality is more appealing because of the edge-orientation property. The experimental results and analysis show that proposed scheme provides a good combination of visual quality and time complexity.

Meat and fish freshness inspection system based on odor sensing.

We propose a method for building a simple electronic nose based on commercially available sensors used to sniff in the market and identify spoiled/contaminated meat stocked for sale in butcher shops. Using a metal oxide semiconductor-based electronic nose, we measured the smell signature from two of the most common meat foods (beef and fish) stored at room temperature. Food samples were divided into two groups: fresh beef with decayed fish and fresh fish with decayed beef. The prime objective was to identify the decayed item using the developed electronic nose. Additionally, we tested the electronic nose using three pattern classification algorithms (artificial neural network, support vector machine and k-nearest neighbor), and compared them based on accuracy, sensitivity, and specificity. The results demonstrate that the k-nearest neighbor algorithm has the highest accuracy.

Digital image super-resolution using adaptive interpolation based on Gaussian function

This paper presents a new approach to digital image super-resolution (SR). Image SR is currently a very active area of research because it is used in various applications. The proposed technique uses Gaussian edge directed interpolation to determine the precise weights of the neighboring pixels. The standard deviation of the interpolation window determines the value of the sigma ‘σ’ for generating Gaussian kernels. Therefore, the proposed scheme adaptively applies different Gaussian kernels according to the computed standard deviation of the interpolation window. Laplacian is applied to the image generated by the Gaussian kernels to enhance the visual quality of the output image. It has the significant benefit of being isotropic i.e. invariant to rotation. These features of being isotropic not only resemble human visual perception but also respond to intensity variations equally in all directions for any kind of kernel. It highlights the discontinuities of high frequencies in the image generated by the Gaussian kernel and deemphasizes the regions with slowly varying luminance levels. It also recovers the background missing features while preserving the sharpness of the output image. The proposed scheme preserves geometrical regularities across the boundaries and smoothes intensities inside the high frequencies. It also maintains the textures inside geometrical regularities. Therefore, high resolution (HR) images produced by the proposed scheme contain intensity information very close to the original details of the low-resolution (LR) image i.e. edges, smoothness and texture information. Various evaluation metrics have been applied to compute the validity of the proposed technique. Extensive experimental comparisons with state-of-the-art zooming schemes validate the claim of the proposed technique of being superior. It produces high quality at the cost of low time complexity.

Video summarization using a network of radial basis functions

The exponential increase of video data on the internet demands efficient management schemes for storage, retrieval, and indexing. One of the methods for managing this huge volume of video data is video summarization. Video summarization is a method to generate smart versions of the videos for efficient retrieval and browsing on the internet. Key frame extraction is a type of video summarization in which the video contents are represented by salient frames of the video. Most of the applications of key frames are user-centered whereby the key frames are used to assist human users in video browsing. However, most of the existing techniques for extracting key frames do not encompass users’ feedback in the retrieval of key frames. In this paper, we propose a user-centered scheme for extracting key frames. In our scheme, the system parameters are learned at training time in the light of users’ feedback. For successful modeling of human perception of similarity in k-means clustering, a non-linear model based on a network of Radial basis functions is employed to reduce the semantic gap between index features and human perception. Experimental results show that the proposed scheme gives excellent results as compared to some of the other techniques.

Video Stabilization by Detecting Intentional and Unintentional Camera Motions

The omnipresence of handheld video devices has led to a drastic increase in the amount of videos created by non-professional users. The unwanted jitter caused by the unintentional motion of the video cameras is very common in such videos. In this paper, we present a technique for removing the unwanted jitter from the videos. Our technique is based on the estimation of camera motion parameters using optical flow features. The camera motion parameters are then used to differentiate between intentional and unintentional camera motion by detecting sharp changes in collective motion estimate curve. The experimental results indicate that our proposed methodology significantly improves the quality of the videos.

Anisotropic Diffusion based Brain MRI Segmentation and 3D Reconstruction

Abstract In medical field visualization of the organs is very imperative for accurate diagnosis and treatment of any disease. Brain tumor diagnosis and surgery also required impressive 3D visualization of the brain to the radiologist. Detection and 3D reconstruction of brain tumors from MRI is a computationally time consuming and error-prone task. Proposed system detects and presents a 3D visualization model of the brain and tumor inside which greatly helps the radiologist to effectively diagnose and analyze the brain tumor. We proposed a multi-phase segmentation and visualization technique which overcomes the many problems of 3D volume segmentation methods like lake of fine details. In this system segmentation is done in three different phases which reduces the error chances. The system finds contours for skull, brain and tumor. These contours are stacked over and two novel methods are used to find the 3D visualization models. The results of these techniques, particularly of interpolation based, are impr...