|Research Topic||Paper ID|
| A REVIEW: SATELLITE IMAGE CONTRAST AND BRIGHTNESS ENHANCEMENT TECHNIQUES
Jasmeet Kaur,Er. Prachi
Abstract: Satellite images have low contrast and are not of good quality. So in order to improve the contrast and brightness of image, contrast enhancement techniques are used. In this paper, overview of different techniques of image, contrast enhancements are being presented which can be used to improve the brightness and contrast of image. These techniques include HE, GHE, BBHE, DSIHE, DWTSVD,DCT-SVD, CS-DWT-SVD. Even the comparison of techniques is also shown.
| HYBRID COMPRESSION TECHNIQUE USING ENCRYPTION AND DAUBECHIES WAVELET
Sunil Khullar,Er. Nandini Chauhan
|HYBRID ALGORITHM OF PSO AND BCO FOR IMAGE CLASSIFICATION OF NATURAL TERRA FEATURES
Dr. Harish kundra,Er. Anika Rana
|HYBRID APPROACH OF BBO AND GA USING IMAGE CLASSIFICATION OF NATURAL TERRAIN FEATURES
Gurpreet Singh, Er. Gopika Kakkar
|Hybrid Algorithm of BFO and ACO for Image Classification Of Natural Terrain Features
Manpreet Kaur, Er. Harneet Kaur
| HYBRID APPROACH ANN AND FL OF IMAGE CLASSIFICATION USING NATURE TERRAIN FEATURES
Dr. Harish kundra,Er. Mamta
Abstract: Image classification is the most important areas in today’s world. Image classified using different methods. Artificial Neural network (ANN) is the most important application in the field of classification. In the paper image classification based upon the different approaches Artificial Neural network (ANN) And fuzzylogic and its seen that fuzzy logic system is better classification techniques than Artificial Neural network (ANN) .Many various have a benefited with the use of Artificial Neural network (ANN) based model over a traditional statistical models for their image classification .In this study, Artificial Neural network (ANN) was hybridized with fuzzy logic (FL) and a new technique Hybrid Artificial Neural network (ANN) to extract natural terrain features was proposed.A simulation tools carry out in the MATALB environments
|Hybrid approach CS and ABC of Natural Terrain Features using Image Classification
Er. Shahida khan,Vivek Sharma
Abstract : This research is for hybridizing two well-known optimization algorithms that provides far much better classification results than the mundane image classification and optimization algorithms. In this research Cuckoo search (CS) and Artificial bee colony (ABC) inspired by the nature. Cuckoo Search algorithm relies on levy flight and on the behavior of cuckoo birds, Besides Artificial Bee Colony algorithm based upon the swarms intelligence , and based upon the bee colony maintain their lives in that colony. Bee’s characteristics are the important part of this algorithm. So in research, hybrid algorithm of Cuckoo Search (CS) and Artificial Bee Colony (ABC) is performed on satellite images to get the better classification results for the images. The key role of using the concept is that the search strategy conducted by Cuckoo search is replaced by ABC algorithm for finding the best host nest required for cuckoo egg and hence cuckoo will place it eggs in those nests. For the research multi-spectral, high resolution satellite image of Alwar region has been taken as input for hybrid algorithm and the results produced by the hybrid algorithm are compared with the result obtained by other swarm intelligence based Hybrid algorithm rough/BBO, FPAB/BBO, fuzzy/BBO, ACO/BBO, CS/ACO CS/PSO to show the evidences of our proposed hybrid algorithm
| Vehicular Ad Hoc Communication Security Enhancement
Deepika Sharma, Er.Sushil Lekhi
Abstract: The large and rapid changes that know all the domains in the world not excluded the transport sector. Today, the fleet is growing; the roads are becoming more dangerous by the effect of congestion and increase the likelihood of collusion. Therefore, securing traffic becomes not only a necessity but also an obligation. The key challenge in Vehicular communications is how to combat with the high mobility of vehicles due to their different speed, as they communicate with each other via the Access Point (AP). Vehicles moving at high speeds have short opportunity to distribute data among each other, and this has to be within the shortest time; else there will be collision. . In this paper objective of our work is to make the secure communication possible in the VANETs environment. Various existing security mechanisms are already proposed but they need the sharing of private key which is not thought to be secure and efficient as it may need frequent change of private key to keep the communication visible only to authorized entities. To show our secure model we will simulate the VANET environment on NS-2 simulator. The network will be consisting of a number of RSUs and moving OBUs. The communication will be designed securely in C++.
| PERFORMANCE ANALYSIS OF PEGASIS PROTOCOL WITH GENETIC APPROACH IN WSN Enhancement
Narottam Saini, Amanpal Singh Rayat
Abstract:Wireless sensor network is turning into a dynamically imperative and testing exploration territory. Headway in WSN empowers an extensive variety of natural observing and item following framework. Remote sensor systems comprise of little minimal effort sensor hubs, having a restricted transmission reach and their preparing, stockpiling abilities and vitality assets are constrained. We consider vitality compelled remote sensor system conveyed over a district. The fundamental assignment of such a system is to assemble data from hub and transmit it to base station for further handling. For the most part, it needs a settled measure of vitality to get one
| Rotation of Motor Through Coin Detection
SWATI BISHT, SRISHTI ARORA
Abstract:Coin Detection is used for differentiating coins on the basis of different coin denominations by applying multiple algorithms. As the coins we use earlier have different feature from that which are being used now. So these features help in classifying different coins. Coins in India are classified on the basis of different parameters like some of the parameters on which we can differentiate coins from each other are such as shape, size, surface design, and weight and so on. Therefore, it is easy for the automated machine to classify Indian coins. Different coin recognition approaches have been used but the most successful method for coin detection is Image recognition. In this paper we proposed an algorithm which demonstrate the coin detection technique in which we assign some values to different coins and then match them with the values stored in the blob , if the value matched we rotate the motor else check again for new values.
| SOFTWARE FAULT INVESTIGATION USING K MEANS
Simerjeet kaur, Anantdeep kaur
Abstract: Software maintenance has been a necessity for the smooth running of the system. Large amount of data is available in different modes like videos, images etcetera. It has become very difficult to manage it time to time. Therefore, Cluster testing had been used in different ways to detect the faults in system. Moreover, Cluster testing works by making the clusters of data and perform testing that make it more effective and prompt. The current work is done to examine the effectiveness of different clustering algorithms such as k means algorithm, density based clustering, quad tree based clustering etcetera and to use the effective one to eliminate the problem of timely detection of errors in the software system.
| A Novel Hybrid Approach of Fuzzy Logic and AOSEDV for the Detection of Sleep Deprivation Attack in MANET
Divjot Kaur, Er.Sushil Lekhi
Abstract: In this research work, we focus on the detection and prevention methods of sleep deprivation attack. The victim of this attack is a battery powered computing device, such as a sensor node, which attempts to remain in a low power sleep mode for as long as possible without adversely affecting the nodes applications. Most of the existing works on sleep deprivation attack detection involve a lot of overhead, leading to poor throughput. Certain level of security can be obtained from the existing solutions. However, these solutions are not always necessarily suitable for wireless networks. In this paper, we have proposed a hybrid concept of AOSEDV and Fuzzy logic for the detection of sleep deprivation attack. The obtained results are compared with AODV protocol by considering the parameters of throughput, packet delivery ratio and path optimality.
| A Novel Hybrid Approach of Neural Network and AOTDV for the Detection of Sybil Attack in Ad-hoc Network
Gurjeet Kaur, Er.Sushil Lekhi
Abstract: Recent advances in information technology has made it possible to transmit the data in wireless -links without the aid of any fixed infrastructure or centralized administrator. Wireless ad hoc networking is a technology that allows fast, easy and inexpensive network deployment. Unfortunately, the advantages of wireless networks also make the task of an attacker simpler, as it is also becomes easier to deploy a malicious node in the environment. There are many attacks which ruin the communication among the nodes of network. Among those attacks there is a Sybil attack that causes severe hazards to the network. Here, we focus on the detection and prevention methods of Sybil attacks on the ad hoc network. In a Sybil attack a malicious node can generate and control a large number of logical identities on a single physical device. The identities used by Sybil attackers are either created by it or uses someone else’s identity. In this paper, we are applying the integrated concept of Neural Network for the Sybil attacks detection and AOTDV (Ad-Hoc on demand trust based Distance vector) Protocol in multi routing using Link Share Technique to prevent these fake identity attacks.
| Enhancement of Satellite Images by DWT-SVD based upon BFO Technique
Pooja Rana, Vinay Chopra
Abstract: The satellite images are of low quality mostly and they are degraded. So they need to be enhanced. The proposed methodology is using DWT-SVD based upon BFO for enhancing the satellite images. BFO is Bacterial Foraging Optimization. The proposed methodology is compared with the existing methodology in terms of PSNR, MSE, BER and MAE .The results show that the proposed methodology is better than the existing technique.
| Software Fault Analysis using k means Algorithm
Simerjeet kaur, Anantdeep kaur
Abstract: Software maintenance has been a necessity for the smooth running of the system. Large amount of data is available in different modes like videos, images etcetera. It has become very difficult to manage it time to time. Therefore, Cluster testing had been used in different ways to detect the faults in system. Moreover, Cluster testing works by making the clusters of data and perform testing that make it more effective and prompt. The current work is done to examine the effectiveness of k means algorithm for cluster testing and to use it to eliminate the problem of timely detection of errors in the software system.