This paper aims at improving quantitative measures of emphysema in computed tomography images of lungs. Pulmonary Emphysema is a chronic obstructive lung disease which is characterized by limitation of air flow. Detection and Quantification of emphysema is important, since it is the main cause of shortness of breath and disability in Chronic Obstructive Pulmonary Disease. Current standard measures like Relative Area method rely on a single intensity threshold on individual pixels, ignoring any interrelations between pixels. HRCT is a sensitive method for diagnosing emphysema, assessing its severity, and determining its subtype. In this project texture based classification system is used. A study is \nperformed by choosing Local binary Pattern (LBP) as texture feature and classification is performed using KNN classifier. Classification accuracy and quantification shows that KNN classifier performs better when LBP is used as texture feature in determining Pulmonary Emphysema.

Wireless Sensor Networks (WSNs) are susceptible to various attacks as they are placed in hostile environments. Several security and authentication mechanisms are proposed for in terms of key exchange mechanisms, handshake protocols, and other routing protocols. The proposed triple key based broadcast authentication protocol is compared with the several other existing security schemes in WSN. The proposed triple key based broadcast authentication scheme works upon TESLA (Timed Efficient Stream Loss-tolerant Authentication) protocol and ECDH (Elliptic Curve Diffie-Hellman) key agreement scheme. The proposed WSN authentication performs better compared to other security schemes, in terms of accuracy, detection of attacks, resiliency, memory consumption, nodal detection, and average of total transmission energy consumed per node.

Trail and punishment of the head of states that commit international crimes were faced with many impairs\nand problems. Among these, the most important legal impair is criminal immunity. The history has shown that these\naccused have generally had s sufficient immunity and not ever s submit to justice. Nullification of the criminal\nimmunity of the head of states is necessary because of matters including public international order, international\npeace and security, domestic international order, prevention of non-punishment of the criminals and deterrence so\nthat prevent of non-punishment of the criminals. Effective implementation of prosecution and trail of the head of the\nstates and nullification of their immunity deals with some impairs, constrains, costs and challenges. Among these\nimpairs are the non-existence of necessary and suitable statutes, non-willing of states and marshaling evidence.\nThese impairs should be identified and rectifies by states. This paper is designed in two parts. The first one deals\nwith necessities of nullification of the criminal immunity and the second part analysis the impairs and constrains of\npunishment of the head of states.

Speaker recognition systems are some kind of biometric systems used in security applications based on remote control phones. This paper presents an approach to balancing the increase in the number of speakers, runtime and error rate of the system. The efficiency of these systems often depends on a particular number of people.With a sudden increase in the number of speakers in terms of time, the recognition error rate experiences a sharp rise. This paper introduces a parameter to select the optimal number,noisy samples for training and testing the Support Vector Machine (SVM) classifier.This approach aims to significantly increase learning speed andto test speaker recognition systems by setting noisy parameters for learning the model and recognizing people. In the first stage,we use the ReliefF algorithm to select the optimal feature subset, then the SVM classifier to achieve better accuracy in recognition by changing existing SVM kernels and automatically setting the balance parameter during training. This method is evaluated based on the accuracy/(Total time ) ratio. The results show that by setting the balance parameter compared with non-setting, above ratio always shows a higher value.

A mobile ad hoc network (MANET) consists of a set of mobile hosts that can communicate with each other without the assistance of base stations. In MANETs, the dynamic nature of the network topology is a major reason for link failures. In this paper, we proposed the Ad hoc on-demand distance vector (AODV) routing (FLAODV) protocol for routing in mobile ad hoc networks to increase durability path. The performance of this routing protocol is studied using opnet10.5. The simulation results of the FLAODV show that the protocol is quite efficient and superior to AODV with the respects to the average Route Discovery Time, the average Packets overhead, average throughput and average end-to-end delay.

Ad hoc networks are dynamically configurable wireless networks that have no fixed infrastructures and do not require predefined configurations. \nIn large, distributed systems, like ad hoc networks, centralized learning of routing or movement policies may be impractical. We need to employ learning algorithms that can learn independently, without the need for extensive coordination. A search for alternative methods of routing packets has resulted in reinforcement learning (RL) as a good approach to adaptive routing. RL methods are able to learn and adapt to a unknown and changing environment. In This paper we propose a new intelligent routing protocol with RL approach for wireless ad hoc networks. The Intelligent Wireless Ad Hoc Routing protocol (IWARP) is an on-demand protocol and self configuring. It selects optimal routes based on local information and past experience.\nThe proposed scheme uses the distributed Q-Learning framework, to select a stable route in order to enhance system performance. Our study also compares the performance of the IWARP protocol with the well-known Ad hoc On-Demand Distance Vector (AODV) protocol, Results obtained by a simulation campaign show that IWARP increases the throughput and decreases the data dropped and number of hops per route.