Urban village recognition and population estimation method and system based on deep learning and computer readable storage medium

Abstract

The invention provides a village-in-city identification and population estimation method based on deep learning and a system-level computer readable storage medium, the method comprises a village-in-city identification stage and a village-in-city population estimation stage, the village-in-city identification stage comprises the following steps: extracting an urban road network map, extracting a road network contour on the road network map by using an opencv python packet, cutting out image blocks on the remote sensing satellite image; carrying out urban village labeling on the cut image blocks, selecting samples to form a training sample set, and carrying out training and prediction by using a Mask-RCNN model to obtain an urban village distribution map on the urban remote sensing satellite image; in the village-in-city population estimation stage, each village-in-city remote sensing satellite image is cut from a village-in-city distribution map on the urban remote sensing satellite image, and house capacity characteristics, crowd movement characteristics and regional function characteristics are obtained through calculation. And the urban village population quantity is trained andpredicted by taking the three as input and utilizing a residual network model. The method provided by the invention has the advantages of high efficiency and low consumption, and achieves higher recognition and estimation accuracy.

Description

technical field [0001] The present invention relates to the field of urban computing, in particular to a method, system, and computer-readable storage medium for identifying and estimating urban villages based on deep learning. Background technique [0002] Urban villages appear in the process of rapid urbanization in many developing countries. They are residential areas that lag behind the pace of urban development, dissociate from modern urban management, and have low living standards. They pose serious social and economic challenges to urban development. Locating and dividing urban village areas and counting the population are very important for the government to renovate and plan urban villages. [0003] Traditional urban village identification and population estimation methods mainly rely on field surveys and population censuses by urban managers. In order to clearly delineate the boundaries of urban villages, it is necessary to ensure that investigators have a compreh...

AI Technical Summary

Problems solved by technology

The disadvantage of this traditional method is that the investigation process causes a lot of time and human and financial resources.

Images