In remote sensing, change detection is used in land use and cover analysis, forest or vegetation assessment and, flood monitoring. Although manual change detection is an option, the time required for it can be prohibitive. It is also highly subjective depending on the expertise of the analyst. Hence, the need for automated methods for such analysis tasks have emerged and thus gives rise to unsupervised machine learning algorithm. This paper analyzes the effectiveness of the three types of unsupervised Machine learning algorithms (MLAs) for change detection to detect the change in some of the dominant classes in an urban area, such as, vegetation, built-up and water bodies. Landsat 5 TM and Landsat 8 OLI imageries have been selected for a part of New Delhi and its nearby area. In this study, three indices namely Normalized Difference Built-up Index (NDBI), Modified Normalized Difference Water Index (MNDWI) and Modified Soil Adjusted Vegetation Index (MSAVI₂) have been generated from the Landsat data. Three algorithms, namely, K-Means, FCM and EM have been used, since these represent three different concepts in machine learning category i.e. partition based, fuzzy and probability based respectively. The same have been implemented in MATLAB for identifying different type of land cover over a period of 1998-2011. Considering both the intra- and inter-cluster distances, silhouette coefficients have been used for evaluation of cluster quality. The change is quantified in terms of percentage that depends upon the outcome of clustering and the number of pixel grouped in each class i.e. urban, vegetation and water.

Sign language is a visual language that uses hand motions, changes in hand shape, and track information to convey meaning. It is the primary mode of communication for those with hearing and language impairments. The use of sign language for communication is limited, despite the fact that sign language recognition can help a large number of such persons deal with regular people. As a result, there is a need to create a more comfortable approach for people with hearing and language impairments to learn and work in order to improve their lives. Therefore, the basic idea behind this article is to make the communication between normal human beings and deaf people much easier. In order to recognize static gestures associated with sign language alphabet and a few commonly used words, we conducted a comprehensive research study employing the hand tracking technique Mediapipe and a gesture classification model based on Support Vector Machine (SVM). The results of the experiments are validated using Recall, F1 Score and Precision. Based on the validated results, we recommend the application of the discussed techniques for such communication. The suggested methods have high generalization qualities and deliver a classification accuracy of around 99 percent on 26 alphabet letters, numerical digits, and some regularly used words.