The Image Recognition Based On Knowledge And The Applied Research In Planning Enforcement

With the rapid development of society, the pace of urbanization has gradually accelerated, however, the rapid urban construction and development bring many problems to the urban planning and management. In recent years, the rapid development of remote sensing and geographic information system technology provide the new methods and new perspective for urban planning and management to address the current problems. It helps to increase efficiency, accuracy of urban planning and provide effective technical means for urban management.In cities, artificial surface features is an important element in the Urban Spatial Information Library,Which including buildings, bridges, roads and buildings and other major projects. According to statistics,80% of the goal is buildings and roads in the High Resolution Image of the Urban areas. Recognition and Extraction of buildings account for a large proportion of Image recognition in the city. How to identify and extract the buildings in the remote sensing images is an important research topic in Remote Sensing Image Recognition.In the process of building in the city, a number of illegal construction which does not meet the urban planning started to appear. Investigate and deal with illegal construction are inspected by law enforcement officials or reported by the masses in previous which has Great chance and bring a great deal of difficulty for the urban planning law. Identification and extraction of the building used in urban planning provides a good opportunity. Knowledge-based Automatic Identification Technology for Image can quickly and interpretation of a wide range of target objects.It solves the problem that the traditional method of manual interpretation of the target image is difficult to timely treatment.This work is divided into the following two parts:1 prosessing the remote sensing satellite data of different time, building a knowledge base for man-made features, through the use of top-down knowledge (assumptions)-driven strategy. Corresponding algorithm designed to deal with the object recognition process, the main follow the typical low, medium and high three-treatment, respectively, image segmentation, feature extraction and model validation.2 Comparing the different periods of urban images, which through the automatic extraction of image feature elements To discover areas of urban change and complete change detection. Analysis of the legality of the building area of urban change with the urban planning data. assisted the law enforcement officers of the urban planing to definite the planning of the urban illegal construction, and fast and accurate for law enforcement. Knowledge-based image recognition technology to detect illegal construction, improve the illegal construction of the "discovery mechanism".It can achieve a wide range, visualization, dynamic monitoring of short-period, and provides technical support for planning and effective law enforcement.