Valliappan Raman

Internet of Things(IoT) digital forensic investigation model: Top-down forensic approach methodology

The Internet of Things (IoT) is the interconnection of uniquely identifiable embedded computing devices within the existing Internet infrastructure. Typically, internet of things (IoT) is expected to offer advanced connectivity of devices, systems, and services that goes beyond machine-to-machine communications (M2M) and covers a variety of protocols, domains, and applications. The interconnection of these embedded devices including smart objects, is expected to usher in automation in nearly all fields, while also enabling advanced applications like a Smart Grid. The main research challenge in Internet of things (IoT) for the forensic investigators is based size of the objects of forensic interest, relevancy, blurry network boundaries and edgeless networks, especially on method for conducting the investigation. The aim of this paper is to identify the best approach by designing a novel model to conduct the investigation situations for digital forensic professionals and experts. There was existing research works which introduce models for identifying the objects of forensics interest in investigations, but there were no rigorous testing for accepting the approach. Currently in this work, an integrated model is designed based on triage model and 1-2-3 zone model for volatile based data preservation.

Spam Detection Proposal in Regular and Text-based Image Emails

Spam, Emails are invading users without their consent and filling their mail boxes with Email trash. Priceless effort and time of the users and organizations are wasted in handling them. To circumvent anti-spam solutions, many spammers are sending spam Email with image-only content. In this paper we propose a spam detection approach in Emails with text and image contents. In the first part, a novel framework for extracting intelligent information from Emails with image content is presented and a prototype implementation is shown. In the second part, a proposal for multi-layered spam detection algorithm is presented, which enhances existing approaches

Matab Implementation and Results of Region Growing Segmentation Using Haralic Texture Features on Mammogram Mass Segmentation

In digital mammograms, accurate segmentation of tumor is very important stage; therefore we have chosen region based segmentation using haralic texture feature for our research. The main objective of this paper is to identify regions of interest and segment the tumors in digital mammograms. The segmented image is then analyzed for estimating tumor and the results are compared against previously known diagnosis of the radiologist. This paper shows the matlab implementation and experimental results of various stages in detecting and segmenting the tumor.

Review on Mammogram Mass Detection by Machine Learning Techniques

Breast cancer continues to be a significant public health problem in the world and number one cause for death rate in Malaysia. Early detection is the key for improving breast cancer prognosis. Mammography is the most effective tool now available for an early diagnosis of breast cancer. However, the detection of cancer signs in mammograms is a difficult task due to irregular pathological structures and noise which are present in the image. It has been shown that in current breast cancer screenings 8%-20% of the tumors are missed by the radiologists. For this reason, a lot of research is currently being done to develop systems for computer aided detection to improve the accuracy. In this paper, review of mammogram mass detection and segmentation is focused. The main aim of the paper is to summarize and compares the method of mass detection in mammogram images. In specific, preprocessing, segmentation, feature extraction and classifications are discussed, Receiver operating curve and free-response receiver operating curve of each method is highlighted to show the sensitivity and specificity of the tumor detection.

Digital Mammogram Tumor Preprocessing Segmentation Feature Extraction and Classification

Mammography has been one of the most reliable methods for early detection of breast carcinomas. The main objective of this paper is to detect and segment the tumor from mammogram images that helps to provide support for the clinical decision to perform biopsy of the breast. In this paper, there are two aspects to segmentation in mammography. First is to separate out the mammogram from the background and the identification of putative masses and the pectoral muscle. The extraction approach is done using basic region growing method to identify the tumor. Second is to extract the features from segmented masses and classifies the masses by case base reasoning method. The experimental results are shown in this paper till the first phase of mass segmentation.

Computer Aided Legal Support System: An initial framework for retrieving legal cases by case base reasoning approach

Law is largely about cases, it is particularly an interesting domain for CBR researchers. The paper explains about how lawyers retrieve information in the process of legal research and reasoning. CALLS (computer aided legal support system), is a reasoning method for determining similar cases by executing a case-similarity calculation step and utilizing a distribution of values of fields for calculation of reasoning results of cases having high similarities. In this paper we provide an overview of our case base reasoning approach in legal domain and outlined the initial frame work of analyzing and retrieving the cases. We discuss the key components of CALLS, in general and the correspondence of those elements in CALLS. The domain for investigation through out the paper is the law of negligence. Finally we concluded the paper with worked through example.

Performance based CBR Mass detection in mammograms

Breast cancer continues to be a significant public health problem in the world. Early detection is the key for improving breast cancer prognosis. Mammography has been one of the most reliable methods for early detection of breast carcinomas. However, it is difficult for radiologists to provide both accurate and uniform evaluation for the enormous mammograms generated in widespread screening. The main objective of this paper is to enhance, detect and classify masses in digital mammogram. We develop a performance based case-based reasoning classification algorithm for mammographic findings to provide support for the clinical decision to perform biopsy of the breast. The developed classifier will be used for training and testing the images which is cancerous and noncancerous and improve the performance of the system.

Cardiovascular diagnostics mining on Borneo island: From labour-intensive to prototype

Mining for medical data poses different challenges compared with mining other types of data. The wide range of imaging modalities of medical data leads to data integration and compatibility issues. The analysis of imaging modalities is further complicated by the different format and attributes used by the different imaging equipment by different vendors. Human factors such as interest of adapting data mining into diagnosis and planning process raised the difficulty of engaging the users into the development of a practical and useful data miner. Requirement engineering technique prototyping further enhanced the engagement of users towards the data-miner. Data from different equipment and different vendors are also merged for efficient data analysis and subsequently charting and reporting. We have also successfully engaged the medical doctors into believing the data miners capability after they reviewed and walkthrough the prototype.

First Prototype of Aquatic Tool Kit: Towards Low-Cost Intelligent Larval Fish Counting in Hatcheries

The demand for aquaculture tools have been increasing due to the rising needs for fish. Several factors show the potential requirements for software solutions. For instance, the continual market value growth of fish, government focus on innovative technologies and research on increasing fish production. The Aquatic Tool Kit aims to calculate the total number of larval and juvenile fish through image processing techniques. Furthermore, it intends to address existing issues during fish counting which has lead to poor accuracy rates and high margins of error. Edge detection and basic morphology tools are utilized to segment and identify fish larvae from acquired images. A preliminary experiment shows a potentially high accuracy rate when detecting larvae and ants.

Cancelable iris Biometrics based on data hiding schemes

The Cancelable Biometrics is a template protection scheme that can replace a stolen or lost biometric template. Instead of the original biometric template, Cancelable biometrics stores a modified version of the biometric template. In this paper, we have proposed a Cancelable biometrics scheme for Iris based on the Steganographic technique. This paper presents a non-invertible transformation function by combining Huffman Encoding and Discrete Cosine Transformation (DCT). The combination of Huffman Encoding and DCT is basically used in steganography to conceal a secret image in a cover image. This combination is considered as one of the powerful non-invertible transformation where it is not possible to extract the exact secret image from the Stego-image. Therefore, retrieving the exact original image from the Stego-image is nearly impossible. The proposed non-invertible transformation function embeds the Huffman encoded bit-stream of a secret image in the DCT coefficients of the iris texture to generate the transformed template. This novel method provides very high security as it is not possible to regenerate the original iris template from the transformed (stego) iris template. In this paper, we have also improved the segmentation and normalization process.