No abstract

No abstract

The proposed system for voting in this project provides a decentralized platform on which all relevant voting information will be available to the public; therefore, security and transparency are the primary objectives of this project. This system uses facial recognition which helps the Election Commission to avoid rigging, bogus results, disputes, and uncertain situations. It uses artificial intelligence for facial recognition of the voters by validating it with a dummy database in which complete detail of a person is available including name, father's name, id number, thumb impression, photo, etc. Then voters can cast their vote easily. Our project infrastructure includes the admin panel which requires login credentials. The system is restricted and cannot be accessed outside of the polling station. In the polling station, firstly, the voter has to input their legal details (NIC) for authentication which redirects the voter to the Facial Recognition window, where the voter's face is validated through this method then all the candidates’ lists will be displayed on the screen of the voter. The program keeps the vote count of every candidate on submission of each vote and displays the result of the election as soon as the voting process ends, which means no delay in results. As our system is a Decentralized system, it also sends the vote count to every node in an encrypted way.

In this research, image processing is used to enhance the traditional way of paddy diseases detection which is done through manual observation by the farmers at the paddy field. Manual observation by farmers requires deep knowledge and full understanding on recognizing the paddy diseases. The knowledge in detecting the paddy diseases is gained through years of observation and experience. Therefore, an appropriate technique is proposed and expected to assist the farmers in detecting the paddy diseases. This paper discussed the method on capturing the sample images using drone, and a technique used in image processing to detect Bacterial Leaf Blight (BLB) disease. The Hue, Saturation, Value (HSV) Band Threshold method consists of pre-processing process, HSV band threshold estimation value, masking, and morphological operation process. In estimating the HSV band threshold, the

No abstract