An object-based image analysis approach for determining the pattern of urban growth in the first planned city of India

Abstract

The world is undergoing the most significant wave of urban growth in history. It is expected that by 2030 the number of people living in the cities will increase to about 5 billion. The rapid urbanization has led to complex problems, including a reduction in vegetation cover, the formation of the urban heat island effect, environmental pollution, reduced open space, etc. This study intends to explore the spatial patterns of urbanization and its impact on the environment in and around Chandigarh- the first planned city of India. Chandigarh was originally planned for a population of 5 lakh, but the city has expanded rapidly over the last four decades and faces problems common to other growing cities in India, including the proliferation of slums and squatter settlements. The areas adjacent to the city boundary also face similar issues. The study presents the methods and results of an object-based classification and post-classification change detection on multi-temporal Landsat data (1978-2017). The processed data was used as an input for object-based classification using image segmentation algorithm of eCognition Developer software. The results show that maximum urbanization has taken place in the last decade in the southern and north-western directions outside the city as a result of the development of an international airport, new sectors and approach roads on the vegetated areas. As a result, maximum changes could be seen in the class vegetation as it has been rapidly changed to built-up areas. The results of this kind of study may hold immense value for planning the urban sprawl areas where up-to-date information is lacking because of the rapid pace of development.