Role of computers in material science and design of classification model to search for the vacancy in outdoor parking lots. (2020)

• Record Type: Journal Article
• Authors: Mago, Neeru; Kumar, Satish
• Is Part Of: Materials today. Volume 28:Part 3(2020)
• Page Start: 1376
• Page End: 1381
• Publication Date: 2020
• Subjects: Machine learning -- Artificial neural network -- K nearest neighbour -- Naive Bayes classifier -- Support vector machine -- Parking system; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2020.04.804
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: With the rapid increase in technology, the role of computers in material science and engineering is growing gradually. For example, database, data processing, mathematical model and simulation techniques can be used in order to utilize technology effectively. Computer data analysis process can be used in materials science for establishing the mechanism model. Many technical problems in materials science and engineering can be solved by computer science technology. The day is not far away when we will overcome the problem of finding vacant lots in outdoor parking area at entry level. Since the vehicles on roads are increasing at exponential rate, the drivers find it difficult to find vacant parking lots in outdoor parking areas. We all must have experienced such situations in our routine life during searching for vacant slots in a shopping mall, multiplexes or outside any building, especially in urbanized environment. This makes it necessary to find an optimized technique to resolve the issues regarding car-parking. In simple words, an efficient technique should be developed to find the status of outdoor parking lots. In this paper, we have conducted experiments on standard dataset "PKLot" consisting of 695, 899 images of two different parking lots captured from three different cameras containing images under different weather conditions like sunny, cloudy and rainy. We have trained and tested various machine learning techniques on the dataset and compared their experimental results.