Ieee Staff

Bio: Ieee Staff is an academic researcher. The author has an hindex of 1, co-authored 1 publications receiving 18 citations.

Medical Image Encryption: An Application for Improved Padding Based GGH Encryption Algorithm

Abstract: Medical images are regarded as important and sensitive data in the medical informatics systems. For transferring medical images over an insecure network, developing a secure encryption algorithm is necessary. Among the three main properties of security services ( i.e. , confidentiality, integrity, and availability), the confidentiality is the most essential feature for exchanging medical images among physicians. The Goldreich Goldwasser Halevi (GGH) algorithm can be a good choice for encrypting medical images as both the algorithm and sensitive data are represented by numeric matrices. Additionally, the GGH algorithm does not increase the size of the image and hence, its complexity will remain as simple as O(n2 ). However, one of the disadvantages of using the GGH algorithm is the Chosen Cipher Text attack. In our strategy, this shortcoming of GGH algorithm has been taken in to consideration and has been improved by applying the padding (i.e., snail tour XORing), before the GGH encryption process. For evaluating their performances, three measurement criteria are considered including (i) Number of Pixels Change Rate (NPCR), (ii) Unified Average Changing Intensity (UACI), and (iii) Avalanche effect. The results on three different sizes of images showed that padding GGH approach has improved UACI, NPCR, and Avalanche by almost 100%, 35%, and 45%, respectively, in comparison to the standard GGH algorithm. Also, the outcomes will make the padding GGH resist against the cipher text, the chosen cipher text, and the statistical attacks. Furthermore, increasing the avalanche effect of more than 50% is a promising achievement in comparison to the increased complexities of the proposed method in terms of encryption and decryption processes.

A novel approach to the design of a new hybrid security protocol architecture

Abstract: This research paper is simply the gathering of recent developments in the field of Hybrid cryptography and its application in the designing of a hybrid security protocol for online transaction based on Hybrid cryptography. A new security protocol for on-line transaction can be designed using combination of both symmetric and asymmetric cryptographic techniques known as Hybrid cryptography. This protocol serves three very important cryptographic primitives - integrity, confidentiality and authentication. Each of the so called cryptographic primitive is provided or fulfilled by the particular symmetric or asymmetric cryptographic techniques. The symmetric cryptographic algorithms are fast as compared to asymmetric cryptographic algorithms, so when both symmetric and asymmetric algorithms are used in tandem or together in a proper way, then the result is very encouraging in terms of providing high security with fast speed. With the help of this survey paper, we tried to encapsulate all the developments introduce in the designing of new security protocol for On-line transaction and its importance is very much evident from the fact that Communication has a major impact on today's business. It is desired to communicate data with high security and in less amount of time. At present, various types of cryptographic algorithms provide high security to information on controlled networks. These algorithms are required to provide data security and users authenticity. A Hybrid security protocol has been designed for better security using a combination of both symmetric and asymmetric cryptographic algorithms.

A Comparative Study on Machine Learning Algorithms for the Control of a Wall Following Robot

Abstract: A comparison of the performance of various machine learning models to predict the direction of a wall following robot is presented in this paper. The models were trained using an open-source dataset that contains 24 ultrasound sensors readings and the corresponding direction for each sample. This dataset was captured using SCITOS G5 mobile robot by placing the sensors on the robot waist. In addition to the full format with 24 sensors per record, the dataset has two simplified formats with 4 and 2 input sensor readings per record. Several control models were proposed previously for this dataset using all three dataset formats. In this paper, two primary research contributions are presented. First, presenting machine learning models with accuracies higher than all previously proposed models for this dataset using all three formats. A perfect solution for the 4 and 2 inputs sensors formats is presented using Decision Tree Classifier by achieving a mean accuracy of 100%. On the other hand, a mean accuracy of 99.82% was achieves using the 24 sensor inputs by employing the Gradient Boost Classifier. Second, presenting a comparative study on the performance of different machine learning and deep learning algorithms on this dataset. Therefore, providing an overall insight on the performance of these algorithms for similar sensor fusion problems. All the models in this paper were evaluated using Monte-Carlo cross-validation.

A Framework for Assessing Reusability Using Package Cohesion Measure in Aspect Oriented Systems

Abstract: Due to better modularization of crosscutting concerns, the Aspect oriented programming approach enhances the quality of the system as it results in less complex and more readable implementation of the system. As the software applications grow in size and complexity, they require some kind of high level organization. For high level organization of software system, packages are required. A lot of work has been carried out for measuring cohesion in Aspect Oriented Systems (AOS) at class level but very less research has been done for designing package level cohesion metric. Package cohesion metrics plays an important role in analyzing quality of software at package level. According to object oriented design principle, a good software design must have high cohesion with high reusability. Thus a relationship must therefore exist between cohesion and reusability. Number of attempts has been made to evaluate effect of cohesion on external attributes but at class level only. Impact of package level cohesion metrics on reusability for AOS is not yet explored. (a) To implement the proposed package cohesion measure, PCohA, on AspectJ sample packages, (b) to theoretically validate the proposed measure and (c) to find the impact of package cohesion on measuring reusability for AOS. Theoretical validation has been done by proving its validity on four theorems given by Briand et al. For finding the impact of proposed measure on external attributes, correlation has been found between package cohesion, PCohA, and external attribute—reusability. After theoretical validation, it has been proved that the proposed measure is suitable for measuring cohesion at package level. Correlation between package cohesion metric (PCohA) and reusability is calculated by using Karl Pearson Product Moment correlation. The computed values show a strong positive relation between PCohA and Reusability. The proposed package cohesion measure is found to be a useful indicator of external quality factors such as reusability. The proposed metric is also established as a better predictor of code reusability than the existing cohesion measures. The work discussed in this paper can be used for designing high quality software by developing new package level metrics for other quality attributes such as maintainability, changeability etc. as a future work.

Weber Local Descriptor based object recognition

Abstract: Recognizing the objects seems to be the challenging task as the object may be occluded, may vary in shape and position and in size. The proposed method is to recognize objects based on the computation of Weber Local Descriptor as feature to the image patches which are extracted around the salient points over the image in order to represent the local properties of the image. Weber Local Descriptor is applied to the Image Patches in order to extract the Salient features. WLD consists of two parts, one is Differential Excitation and the other is Orientation by which the local salient features are extracted. Then these features are fed to SVM classifier. Benchmark database used here is UIUC database. Experimental results by varying the patch sizes are given and the results obtained are satisfactory.