145 research outputs found
An abandoned object detection system based on dual background segmentation
No full textAn abandoned object detection system is presented and evaluated using benchmark datasets. The detection is based on a simple mathematical model and works efficiently at QVGA resolution at which most CCTV cameras operate. The pre-processing involves a dual-time background subtraction algorithm which dynamically updates two sets of background, one after a very short interval (less than half a second) and the other after a relatively longer duration. The framework of the proposed algorithm is based on the Approximate Median model. An algorithm for tracking of abandoned objects even under occlusion is also proposed. Results show that the system is robust to variations in lighting conditions and the number of people in the scene. In addition, the system is simple and computationally less intensive as it avoids the use of expensive filters while achieving better detection results
Text-independent speaker recognition for Ambient Intelligence applications by using Information Set Features
No full textBiometric systems are enabling technologies for a wide set of applications in Ambient Intelligence (AmI) environments. In this context, speaker recognition techniques are of paramount importance due to their high user acceptance and low required cooperation. Typical applications of biometric recognition in AmI environments are identification techniques designed to recognize individuals in small datasets. Biometric recognition methods are frequently deployed on embedded hardware and therefore need to be optimized in terms of computational time as well as used memory. This paper presents a text-independent speaker recognition method particularly suitable for identification in AmI environments. The proposed method first computes the Mel Frequency Cepstral Coefficients (MFCC) and then creates Information Set Features (ISF) by applying a fuzzy logic approach. Finally, it estimates the user's identity by using a hierarchical classification technique based on computational intelligence. We evaluated the performance of the speaker recognition method using signals belonging to the NIST-2003 switchboard speaker database. The achieved results showed that the proposed method reduced the size of the template with respect to traditional approaches based on Gaussian Mixture Models (GMM) and achieved better identification accuracy
Fusion of hand based biometrics using particle swarm optimization
No full textMulti-modal biometrics has numerous advantages over unimodal\ud
biometric systems. Decision level fusion is the most\ud
popular fusion strategy in multimodal biometric systems. Recent\ud
research has shown promising performance of hand based\ud
biometrics, i.e. palmprint and hand geometry over other\ud
biometric modalities. However, the improvement in\ud
performance is constrained by the lack of optimal sensor points\ud
and fusion strategy. In this paper, we have implemented a\ud
particle swarm based optimization technique for selecting\ud
optimal parameters through decision level fusion of two\ud
modalities: palmprint and hand geometry. The experimental\ud
evaluation on a database of 100 users confirms the utility of the\ud
decision level fusion using particle swarm optimization
Personalization of interactive news through J2EE, XML, XSLT, and SMIL in a Web-based multimedia content management system
No full textThis paper describes the design and implementation of a 5 layered web-based multimedia content management system (MCMS) using the Java 2 Enterprise Edition (J2EE). A prototype based on our framework has been implemented in the News On Demand KIOSK Network for organizing, integrating and composing of personalized digital news for interactive broadcasting. The aim of the MCMS project is to provide a collaborative environment among news producers for making them work more effectively despite the time and location constraints. The MCMS generates SMIL document that is structured, profiled and streamed to end-user using XML and XSLT techniques, which form the backbone of digital news broadcasting. The major contributions with regard to the digital MCMS can be summarized as: (1) Support for effective personalization of multimedia news content and presentation styles through the utilization of XML and XSLT. (2) Separation of design and content facilitated by MCMS. This allows journalist and editors to focus on content preparation rather than advanced HTML and SMIL coding. (3) Support for the re-use and re-purpose operations of the same multimedia elements to be part of the other digital news program. (4) Platform independent MCMS allowing an author to access the application everywhere via Internet without any need of additional hardware or software
Unconstrained handwritten character recognition based on fuzzy logic
No full textThis paper presents an innovative approach called box method for feature extraction for the recognition of handwritten characters. In this method, the binary image of the character is partitioned into a fixed number of subimages called boxes. The features consist of vector distance (gamma) from each box to a fixed point. To find gamma the vector distances of all the pixels, lying in a particular box, from the fixed point are calculated and added up and normalized by the number of pixels within that box. Here, both neural networks and fuzzy logic techniques are used for recognition and recognition rates are found to be around 97 percent using neural networks and 98 percent using fuzzy logic. The methods are independent of font, size and with minor changes in preprocessing, it can be adopted for any language. (C) 2002 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved
Fuzzy model based recognition of handwritten Hindi characters
No full textThis paper presents the recognition of handwritten Hindi Characters based on the modified exponential membership function fitted to the fuzzy sets derived from features consisting of normalized distances obtained using the Box approach. The exponential membership function is modified by two structural parameters that are estimated by optimizing an\ud
objective function that includes the entropy and error function. A Reuse Policy that provides guidance from the past policies is utilized to improve the reinforcement learning. This relies on the past errors exploiting the past policies. The Reuse Policy improves the speed of convergence of the learning process over the strategies that learn without reuse and combined with the use of the reinforcement learning, there is a\ud
25-fold improvement in training. Experimentation is carried out on a database of 4750 samples. The overall recognition rate is found to be 90.65%
FABRIC IMAGE DEFECT DETECTION BY USING GLCM AND ROSETTA
No full textABSTRACT Automated visual inspections of industrial goods for Quality control plays an ever-increasing role in production process as the global market pressures put higher and higher demand on quality at lower cost. In most cases, the quality inspection through visual inspection is still carried out by humans. However, the reliability of manual inspection is limited due to fatigue and inattentiveness. The author did the literature survey on textile industry and the most highly trained inspectors can only detect about 70% of the fabric defects
Color image encryption and decryption using Hill Cipher associated with Arnold transform
No full textImage security over open network transmission is a big concern nowadays. This paper proposes another methodology for color image encoding and decoding using two stage Hill Cipher method which is connected with Arnold Transformation. The forgoing created a strategy for encryption and decryption of color image information and touched on just the premise of keys. In this plan, keys and the agreement of Hill Cipher (HC) are basic. Moreover, keys multiplication (pre or post) over an RGB image information framework is inevitable to know to effectively decrypt the first image information. We have given a machine simulation with a standard example and the result is given to support the stalwartness of the plan. This paper gives a detailed comparison between prior proposed methods and this methodology. The system has potential utilization in computerized RGB image transforming and security of image information
Input Fuzzy Modeling for the Recognition of Handwritten Hindi Numerals
No full textThis paper presents the recognition of Handwritten Hindi Numerals based on the modified exponential membership function fitted to the fuzzy sets derived from normalized distance features obtained using the Box approach. The exponential membership function is modified by two structural parameters that are estimated by optimizing the criterion function associated with the input fuzzy modeling. We then utilize a ‘Reuse Policy ’ that provides guidance from past error values of the criteria function to accomplish the reinforcement learning. We will also show how the ‘Reuse Policy ’ improves the speed of convergence of the learning process over other strategies that learn without reuse. There is a 25-fold improvement in training with the use of the reinforcement learning Experimentation is carried out on a limited database of around 3500 Hindi numeral samples. The overall recognition rate is found to be 95%
- …
