No abstract

Hyperspectral imaging (HSI), also known as imaging spectroscopy or 3D spectroscopy, combines imaging and spectroscopy into a single system. With a high resolution measurement of spectral signatures, HSI is able to provide critical information of the target. Thus it is useful for various scientific and industrial applications, including food safety and disease diagnosis. Due to constantly increasing demands for safe animal products, there is pressure on the processing sector for applications of advanced, high throughput methods for non-destructive quality analysis of animal products. In this context, HSI finds its applications for grading, classification, quality & composition analysis of animal products including meat, egg, milk etc. Further, the technique is also a useful tool in poultry sector for assessment of wholesomeness and quality control of chicken carcasses, as well as, chicken meat products. In fish industry also, the technique has established its potential for determining freshness and quality attributes of sea-foods. Apart from quality control of animal products, HSI has also demonstrated its usefulness for disease diagnosis in animal models and for detection of mammary cancers in dogs. Thus, the future of HSI technology in animal industry is promising and associated with multivariate analysis, HSI technique will further dominate in animal products authentication and analysis in the future also.

Study on geometric properties of nanoparticles and their relation with biomolecular activities, especially protein is quite a new field to explore. This work was carried out towards this direction where images of gold nanoparticles obtained from transmission electron microscopy were processed to extract their size and area profile at different experimental conditions including and excluding a protein, citrate synthase. Since the images were ill-posed, texture of a context-window for each pixel was used as input to a back-propagation network architecture to obtain decision on its membership as nanoparticle. The segmented images were further analysed by k-means clustering to derive geometric properties of individual nanoparticles even from their assembled form. The extracted geometric information was found to be crucial to give a model featuring porous cage like configuration of nanoparticle assembly using which the chaperone like activity of gold nanoparticles can be explained.

Network attacks are becoming more complex, making it more difficult to detect intrusions. Various research have been done over the years, employing different categorization techniques of Data Mining (DM) and Machine Learning (ML) inspired hybrid approaches to develop robust IDS. Almost all researchers suggested to improve accuracy in intrusion detection with low computational cost. Authors observed that dissimilar sets of features were picked for different classifiers to get the highest accuracy. This paper is dedicated to a review of relevant research, where an in-depth investigation was carried out with two emphasis points of IDS, which contain distinct pre-processing techniques in the form of feature selection and a diversity of classification algorithms. In addition, this paper presents a comparative algorithmic assessment of the DM and ML techniques applied to create an intelligent IDS. A novel feature selection method based on the CART algorithm has also introduced which provides optimal feature subset of the dataset so perfectly that it has made various existing DM and ML classifying algorithms more performant than earlier, this makes classifiers independent of feature selection. To validate the performance of proposed work, experiments have performed using the 'Python' programming language and 'corrected' & '10_percent' of 'Kddcup99' datasets used as benchmark. As an outcome proposed work, we found that feature reduction and selecting a classifier had a significant influence on the rate of intrusion detection accuracy. The results of simulation and comparison analysis of proposed work with existing DM and ML based classification approaches show that the suggested work is more competent in true prediction and attaining maximum intrusion detection accuracy with minimal computing cost of prediction.

No abstract