K. K. Biswas

Gesture recognition using Microsoft Kinect

Gesture recognition is essential for human — machine interaction. In this paper we propose a method to recognize human gestures using a Kinect® depth camera. The camera views the subject in the front plane and generates a depth image of the subject in the plane towards the camera. This depth image is then used for background removal, followed by generation of the depth profile of the subject. In addition to this, the difference between subsequent frames gives the motion profile of the subject and is used for recognition of gestures. These allow the efficient use of depth camera to successfully recognize multiple human gestures. The result of a case study involving 8 gestures is shown. The system was trained using a multi class Support Vector Machine.

Region-based image retrieval using integrated color, shape, and location index

A technique to retrieve images by region matching using a combined feature index based on color, shape, and location is presented within the framework of MPEG-7. Dominant regions within each image are indexed using integrated color, shape, and location features. Various combinations of regions are also indexed. The resulting indices and related metadata are stored in a Hash structure, where similar images tend to form clusters. The retrieval process is non-cascading and images can be retrieved based on color, shape or location and also based on a combined color shape location index. Results obtained show that retrieval effectiveness increases in non-cascaded region-based querying by combined index.

Generalized intuitionistic fuzzy soft sets with applications in decision-making

Abstract The concept of intuitionistic fuzzy soft set (IFSS) arising from intuitionistic fuzzy set (IFS) is generalized by including a parameter reflecting a moderators opinion about the validity of the information provided. The resulting generalized intuitionistic fuzzy soft set (GIFSS) finds a special role in the decision making applications. It can evaluate the given criteria along with the moderators assessment of the furnished data. The properties of GIFSS are investigated and the associated relations called generalized intuitionistic fuzzy soft relations (GIFSR) are given. A similarity measure is given to compare two GIFSSs. As this is not applicable to fuzzy numbers, a new score function is devised to compare two intuitionistic fuzzy numbers (IFNs), the components of IFS. The effectiveness of the proposed GIFSS in decision making is demonstrated on four case studies.

A fuzzy theoretic approach for video segmentation using syntactic features

Abstract This paper is concerned with the development of a fuzzy-logic-based framework for segmentation of video sequences. We have proposed a scheme for fuzzification of the frame-to-frame property difference values using the Rayleigh distribution. The difference values have been characterized by fuzzy terms like small, significant, large, etc. These terms have been used to design fuzzy rules for detecting abrupt changes and gradual changes. Fuzzy rules have provided a mechanism for integrating evidences based on different properties. The decompositional inference strategy has been used for fuzzy reasoning over the set of fuzzy rules. Gradual changes have been further classified as fade-in, fade-out and others (including dissolves, wipes, etc). Experimental results have shown that the proposed scheme can detect changes reliably.

Dominant color region based indexing for CBIR

Our world is dominated by visual information and a tremendous amount of such information is being added day-by-day. It would be impossible to cope with this explosion of visual data, unless they are organized such that we can retrieve them efficiently and effectively. The main problem in organizing and managing such visual data is indexing, the assignment of a synthetic descriptor which facilitates its retrieval. It involves extracting relevant entities or characteristics from images as index keys. Then a representation is chosen for the keys and specific meaning is assigned to it. Color is an important cue for content based image retrieval (CBIR) systems. We propose a technique to index and store images based on dominant color regions. Features like region size and location of the region are extracted and used as similarity measures. Images with similar indices are stored as an image cluster in a Hash table. A prototype of the retrieval system is developed using the JAVA language.

Color and Shape Index for Region-Based Image Retrieval

Most CBIR systems use low-level visual features for representation and retrieval of images. Generally such methods suffer from the problems of high-dimensionality leading to more computational time and inefficient indexing and retrieval performance. This paper focuses on a low-dimensional color and shape based indexing technique for achieving efficient and effective retrieval performance. We propose a combined index using color and shape features. A new shape similarity measure is proposed which is shown to be more effective. Images are indexed by dominant color regions and similar images form an image cluster stored in a hash structure. Each region within an image is further indexed by a region-based shape index. The shape index is invariant to translation, rotation and scaling. A JAVA based query engine supporting query-by-example is built to retrieve images by color and shape. The retrieval performance is studied and compared with a region-based shape indexing scheme.

Identification of stochastic time-delay systems

A method is presented to identify time lags present in systems described by linear, stochastic dynamical models usually encountered in process control. The model equation is first discretized and converted into a non-time-delayed form: a two-stage estimator is then proposed to identify the model parameters.

An Approach to Fixed-Point Smoothing Problems

This paper examines the possibility of deriving fixed-point smoothing algorithms through exploitation of the known solutions of a higher dimensional filtering problem. It is shown that a simple state augmentation serves to imbed the given n-dimensional smoothing problem into a 2n-dimensional filtering problem. It is further shown that computation of the smoothed estimate and the corresponding error covariance does not require implementation of the 2n-dimensional filtering equations. Some new results involving systems with or without multiple time delays and having colored observation noise have been derived in order to illustrate the versatility of the proposed technique. It is also demonstrated that the present approach leads to an easier derivation of the continuous-time fixed-point smoothing algorithm reported in the literature.

Optimal fixed lag smoothing for time delayed system with colored noise

The aim of this note is to report an algorithm for the fixed lag smoothing problem of a time delayed system whose observations contain colored noise. This is derived by looking at the appropriate components of a filtering algorithm for a nontime delayed higher dimensional system.

Load modelling for real time monitoring of power systems

The problem of simultaneous prediciton of the load demand at all the (loading) nodes of an interconnected power system is studied. Such predictions are of obvious importance to the practising engineer for real time monitoring of the system and for economic generation scheduling. Existing prediction techniques require the use of a central computer, first for the purpose of model building and then for on-line load prediction. Both these tasks involve the processing of data being recieved from all the nodes of the system. The main purpose of the present study has been to explore the possibility of reducing the computational burden of such an exercise by developing new algorithms. A specific algorithm discussed in the text is based on the use of a two-stage filtering technique in order to replace the nonlinear estimation problem inherent in the model building task by linear estimation problems. After the model is identified, further reduction of computation is shown possible if instead of allthe data, use is made of the load data only from a few important nodes for on-line prediction of the demand at all the nodes. Some results of application of the new algorithm to real data pertaining to the Northern India grid are presented.