REDUCTION OF SEMANTIC GAP USING LOW LEVEL VISUAL FEATURES

¹ A.ATHIRAJA, ² K.BALA MURALI, ³ Dr. A. ASKARUNISA, ⁴ S.UMAMAHESWARAN
¹ Asst Professor CSE, ² PG student CSE, ³ Professor and Head of CSE, ⁴ Asst Professor CSE
Department of Computer Science and engineering, Vickram College Of Engineering, Enathi, Sivagangai, India.

     Digital image libraries and other multimedia databases have been dramatically expanded in recent years. In order to effectively and precisely retrieve the desired images from a large image database, the development of a content-based image retrieval (CBIR) system has become an important research issue. Most of the existing approaches lack the capability to effectively incorporate human intuition and emotion into retrieving images. In order to reduce the semantic gap the proposed approaches emphasize on finding the best representation for different image features. Furthermore, very few of the representative works will consider the user’s subjectivity and preferences in the retrieval process. In this project, a user-oriented mechanism for CBIR method based on low level visual features like color texture, histogram and correlation are used. Color attributes like the mean value, the standard deviation were used. The entropy based on the gray level co-occurrence matrix of an image is considered as the texture feature. Further the histogram values are used for effective image retrieval process; finally the images are compared using correlation values at both the ends. The efficiency of proposed technique is evaluated and the experimental result indicates that it outperforms other existing systems. In this project based on reduction of I have used corel datasets is used which consist of 1000 images of varying 10 semantic contents, out of which 100 images were used.

View full abstract...

COMPUTATION OF IMAGE DISTANCES FOR HUMAN IDENTIFICATION IN DENTAL RADIOGRAPHS

¹ PG Student, ² Assistant Professor, ³ Professor & Head, ⁴ Assistant Professor
Department of Computer Science and engineering, Vickram College Of Engineering, Enathi, Sivagangai, India.

     Dental radiographs are used for human identification in dental biometrics. The dental radiograph gives us various information such as tooth contours, relative positions of neighbouring teeth, and shapes of the dental work (e.g., crowns, fillings, and bridges). The proposed system has 2 stages namely (1) Feature Extraction and (2) Matching. In feature extraction, active contour model is used to extract the contour. The matching stage has 2 steps viz. Computation of Image distances and Subject identification. In tooth level matching tooth contours are matched using “Shape registration Method” and depending upon the overlapping areas the dental works are matched. Then the values of the distance between the tooth contours and dental works are combined using posterior probabilities. Tooth correspondence between query radiograph and database radiograph are established. Distance between the teeth are used to calculate the similarity between the two radiographs. Finally the distance between the radiographs provide the details about the subject associated with these radiographs. The dataset contains 10 normal images and 55 OPG images which was collected from Madura Dental Hospital. The accuracy of the algorithm is measured by the ratio of Correct Detection images to Total No of images. The experimental results show that this proposed algorithm is accurate about 72%.

View full abstract...

CARBON MONOXIDE CONCENTRATION CAUSED BY TRAFFIC IN COIMBATORE CITY USING GEOTECHNIQUES

¹ G.Noushad, ² Dr. A.Ilandhirayan
¹, Assistant Professor, Department of Civil Engineering, Ranganathan Engineering Colleges, Thondamuthur, Coimbatore, Tamilnadu, India.
^2, Assistant Professor, Department of Geography, Karur Govt. Arts College, Karur, Tamilnadu , India.

     The environment has an impact on every aspect of human life and every human activity affects the environment with social, technological and economic development. Abnormal consumption of various natural resources has increased the level of pollutants such as air pollution, chemical pollution, thermal pollution and radioactive pollution etc., which is increasing at an exponential rate, thus endangering our country’s rich ecology and environment. The major sources of air pollutants are man’s industrial manufacturing and motor vehicle operation activities, both of them have been concentrated in urban areas. Study area Coimbatore district is the second most populous city in Tamilnadu and spread over an area of 105.60 sq.km. Study area is well connected with other centres of the country by road, rail and air routes. The National highways are passes through the city. The study area for the present work deals with Carbon Monoxide from vehicular emission due to heavy traffic flow. Vehicular data were collected from 6 sample traffic junctions were selected and the vehicles running time, based on the peak hours of the traffic junctions. Geographic information systems (GIS) as management and supporting systems enable us to integrate and analyze air pollutants on environment. Finally spatial analysis such as overlay, buffering and IDW have been performed studied for carbon monoxide pollution in the city.

View full abstract...

ROLE OF SOCIAL NETWORKS TO LINK PERSONAL HEALTH RECORDS USING ATTRIBUTE BASED ENCRYPTION

¹ C.Kalaivani, ² Mrs.J.V.Anchitaalagammai

¹ PG student- Department of CSE - Velammal college of Engineering and technology, Madurai, Tamilnadu, India.
² Assistant Professor - Department of CSE- Velammal college of Engineering and technology, Madurai, Tamilnadu, India.

     Personal health record (PHR) is a rising patient-centric model of health information exchange which is often stored at a third party. Major problem in this scenario is the privacy of records. To assure the patients control over access to their own PHRs, it is a promising method to encrypt the PHRs before outsourcing. A new cryptosystem for fine grained sharing of encrypted data is developed based on Attribute Based Encryption technique (ABE) in a social networking environment. We propose an assurance of providing security for the PHR records. Not only providing security also provides scalability and flexible access to the records. To achieve fine-grained and scalable data access control for PHRs and to reduce the key distribution complexity, we divide the system into multiple security domains where each domain manages only a subset of the users. In this way, each patient can have full control of their PHR file and the key management complexity is reduced dramatically. Our proposed scheme is also flexible as it supports efficient random-demand revocation of user access rights and break-glass access under emergency scenarios.

View full abstract...

CONTEMPORARY ROUTING CHANNEL HOPPING FOR COGNITIVE RADIO NETWORK

P.Venkatesh

M.E - Department of Computer Science and Engineering Srividhya college of Engineering and technology, Virudhunagar, Tamilnadu, India.

     Recently, cognitive radio (CR) has become a key technology for addressing spectrum scarcity. In CR networks, spectrum access should not interfere the colocate incumbent networks. Due to the requirement above, common control channel approaches, which are widely used in traditional multichannel environments, may face serious CR long-time blocking problem and control channel saturation problem. Although channel-hopping-based approaches can avoid these two problems, existing works still have significant drawbacks including long time-to-rendezvous, unbalance channel loading, and low channel utilization. In this paper, we introduce three channel-hopping approaches, RCCH, ARCH, and SARCH for synchronous and asynchronous environments, respectively. Compared with previous works, our schemes outperform the state of the art in terms of these metrics.

View full abstract...

AN EFFICIENT INDEXING SCHEME TO IMPROVE ACCURACY AND SPEED OF IRIS RECOGNITION SYSTEMS

¹ K.M.Rajshree, ² Ms. G.Pramila, ³ Dr.A.Askarunisa, ⁴ Mr.A.Athiraja
¹ PG Student, ² Assistant Professor, ³ Professor & Head, 4 Assistant Professor
Department of Computer Science and engineering, Vickram College Of Engineering, Enathi, Sivagangai, India.

     In video analysis and image processing, the most successful application of iris authentication. In this project, several novel approaches are proposed to improve the overall performance of iris recognition systems. First, to achieve a low rate of eyelash misclassification, a new eyelash detection algorithm based on directional filter is proposed. Second, fusion method proposes a multiscale and multiorientation data method is introduced to reduce the edge effect, and to increase matching criteria. Finally, to accelerate exhausted the search in a huge iris database performed, using an iris indexing method on the basis of corner detection is presented, and matching process is performed by using hamming distance. The performance evaluations are carried out on two popular iris databases (CASIA IRIS INTERVAL) and (CASIA IRIS TWIN), and the test results are experimentally more accurate with less elapsed time compared with most existing methods.

View full abstract...