Search

Urbanization generally serves as a key navigator of the economic growth and development of the country. There is a need for fast and accurate urban planning to accommodate more and more people in the city area. Remote sensing technology has been used for planning the expansion and design of city areas. A novel machine learning (ML) classifier formed by combining AdaBoost and extra trees algorithm have been investigated for change detection in the urban area of three cities in the Gujarat region of India. Using Indian Remote Sensing (IRS) Resourcesat-2 LISS IV satellite images, the performance of the object-based AdaBoosted extra trees classifier (ABETC) in terms of overall accuracy (OA) and kappa coefficient (KC) for urban area change detection was compared to benchmarked object-based algorithms. As the first step in object-based classification (OBC), the Shepherd segmentation algorithm was used to segment satellite images. For all three cities, the object-based ABETC demonstrated the highest efficiency when compared to conventional classifiers. The rise in the built-up area of Ahmedabad city has been noted by 87.39 sq km from the year 2011 to 2020 showing the urban development of the city. This increase in the built-up area of Ahmedabad was compensated by the depletion of 30.26 sq. km. vegetation area, and 57.13 sq. km. of open land class. The built-up area of Vadodara and Rajkot city has been enlarged by 17.24 sq km and 6.79 sq km respectively. The highest OA of 96.04% and KC of 0.94 has been noted for a satellite image of Vadodara city with a novel object based ABETC algorithm.

In the past decade, it is proven that satellite image classification using an object-based technique is better than the standard pixel-based technique. With the increasing need for classifying multispectral satellite images for urban planning, the accuracy of the classification becomes a significant performance parameter. Object-based classification (OBC) is a technique in which group of pixels having similar spectral properties, called objects, are generated using image segmentation and then these objects are classified based on their attributes. In this paper, the combination of a multiclass AdaBoost algorithm with extra trees classifier (ETC) is proposed with higher prediction accuracy for the OBC of the urban fringe area. The performance of the AdaBoost algorithm is found to be better in terms of classification accuracy than benchmarked SVM and RF classifiers for OBC. These classification methods were applied to IRS-R2 LISS IV data. The AdaBoosted extra trees classifier (ABETC) has demonstrated the highest accuracy with overall accuracy (OA) of 88.47% and a kappa coefficient of 0.85. The computational time of the ABETC is found to be much smaller than the RF algorithm. In detail, the sensitivity of the classifiers was investigated using stratified random sampling with various sample sizes. [ABSTRACT FROM AUTHOR]

In last few years, Docker container is used as light weight virtualization for building infrastructure as a service of cloud computing. Container placement on machines is traditional scheduling problem in docker based cloud. Docker supports default scheduling strategy called spread strategy. Its main objective is to place containers on each and every machine available in docker cluster. However, spread strategy does not consider load of machines. So, it leads to overloading or under loading of machines as well as not proper utilization of resources. In this paper, we proposed new scheduling algorithm that consider CPU usage of every machine in the docker cluster. To choose the most appropriate machine to place containers needed to be allocated in scheduling process, author define method for CPU utilization of every machine and select machine with minimum CPU utilization value for container placement. The experiment results shows that our scheduling algorithm use minimum number of machines in docker cluster as well as proper utilization of resources as compared to spread strategy. [ABSTRACT FROM AUTHOR]