Madhulika Bhatia

An anatomization of noise removal techniques on medical images

In applications of signal processing such as medicine, communications and satellites, preprocessing is considered as a vital step which focuses on reduction or removal of the level of the noise contained in the image. The process of denoising helps in preserving the finer details and useful information. Medical images like MRI, CT and X-ray contain very fine details that need to be correct and free from noise so that the information and features of interests are not lost during the diagnosis. In this paper, various noise reduction techniques such as wavelet transform, Neural Network, PCA, ICA and mean and median filters over medical images has been discussed. In this paper we tried to highlight the strength and weakness of various noise removal techniques over processing of the medical images.

A Cheiloscopic Approach for Unique Identification Among Indian Subpopulation

A biometric is formed on an individual’s behavioural or physical features. The main approach is to uniquely identify humans. Identification by biometric factors finishes the complications related with customary approaches used for human identification. The biometric methods that are most commonly being used today are fingerprints, eye retina, iris, etc. This paper shows that just like fingerprints and lip prints are unique in nature and hence can be used as one of the measures to recognize individuals. Also, this paper shows that the nature of lips of an individual varies according to state one belongs to. The lip print samples are taken from different people in different states. After the enhancement of image, existing Sobel edge detection algorithm has been applied to detect the edges of lips. Thereafter, the numValue, i.e. featureValue of the lip print, is extracted and stored which depicts the uniqueness. The graphs have been plotted and examined.

Anatomization of Software Quality Factors: Measures and Metrics

“If you cannot measure it, you cannot manage it”, this quote by Lord Kelvin is very much applicable to the world of Software Quality as well. Quality of software can be described as the extent to which it performs the task that the user has specified. It can be expressed in terms of multiple factors like reliability, readability, supportability et al. It can be best described as the amalgamation of these multiple factors. Not only the identification of factors but also of the metrics and measures were done by studying and analyzing various research papers and keeping them as the primary foundation. This paper focus on investigating the measures that is already available to determine the different quality factors. The results obtained are advantageous for software developers, researchers and academicians to recognize and distinguish the cadent used to dimension the different quality characteristics of the software. Moreover, the work focuses at giving some suggestions, using the potential deficiencies detected as a foundation.

Enacting Segmentation Algorithms for Classifying Fish Species

The fundamental feature of Computer vision involves consolidating image processing, pattern recognition and classification procedures. Extricating data from a digital picture relies on upon first distinguishing essential objects or dividing the picture into homogenous sectors or objects termed as segmentation and afterward allotting out these sectors or objects to specific classes termed as classification procedure. The term homogeneous may allude to the shade of the region or an object, however it additionally may utilize different characteristics, for example, composition and shape. This study concentrates on implementing image segmentation and classification on six different fish species using the watershed and the nearest neighbor classifier (kNN) algorithm.

IRIS based human identification: Analogizing and exploiting PSNR and MSE techniques using MATLAB

In todays reality, security has gotten to be paramount. Many biometrics methods like facial expression recognition system, Iris recognition system is a standout amongst the most dependable leading technologies for user authentication. It is steady for the duration of the life, it can serve as a living visa or a living secret word that one need not recall however is always present. Iris biometry helps in identifying an individual in a more intuitive and natural manner. Iris recognition focuses on recognizing the identity of individuals using the textural based characteristics of the muscular patterns of the iris. Irises assure long period stability and also infrequent requirements for the enrollment process. Accuracy, higher information content, real timeliness, performance, stability, low circumvention and uniqueness makes iris technology as the one of the most suitable candidate to be deployed in the field of the security. The study attempted to highlight the performance of various preprocessing techniques used in iris recognition in terms of their PSNR values and MSE values.