ABSTRAT:
Road extraction from high-resolution remote sensing images is a challenging but hot research topic in the past decades. A large number of methods are invented to deal with this problem. This article provides a comprehensive review of these existing approaches. We classified the methods into heuristic and data-driven. The heuristic methods are the mainstream in the early years, and the data-driven methods based on deep learning have been quickly developed recently. With regard to the heuristic methods, the road feature model is first introduced, then, the classic extraction methods are reviewed in two subcategories: semiautomatic and automatic. The principles, inspirations, advantages, and disadvantages of these methods are described. In terms of the data-driven methods, the road extraction methods based on deep neural network, particularly those based on patched convolutional neural network, fully convolutional network, and generative adversarial network are reviewed. We perform subjective comparisons between the methods inner each type. Furthermore, the quantity performances achieved on the same dataset are compared between the heuristic and data-driven methods to show the strengthening of the data-driven methods. Finally, the conclusion and prospects are summarized.
EXISTING SYSTEM :
Manuscript received June 17, 2019; revised December 23, 2019, April 7, 2020,and July 28, 2020; accepted September 1, 2020. Date of publication September11, 2020; date of current version September 25, 2020. This work was supported in part by the Natural Science Foundation of China under Grant 61170147, in part by the Education and Scientific Research Project for Middle-Aged and Young Teachers in Fujian Province under Grant JT180595, in part by the Key Program of Fostering Young Scientific Research Talents in Fujian Jiangxi University under Grant JXZ2018002 and Grant JXZ2016001, and in part by the Fuzhou Science and Technology Project under Grant 2020-G-066. (Corresponding authors: Weixing Wang; LiqinHuang.) Renbao Lian is with the College of Physics and Information Engineering, Fuzhou University, Fuzhou 350008, China, with the Digital Fujian, Internet-of-Things Key Lab of Information Collection and Processing in Smart Home, Fuzhou 350108, China, and also with the College of Electronics and Information Science, Fujian Jiangxi University, Fuzhou 350108, China (e-mail:luoshao@163.com
EXISTING SYSTEM DISADVANTAGES:
1.LESS ACCURACY
2. LOW EFFICIENCY
PROPOSED SYSTEM :
Establishing a road model can help us extract road more effectively. Baumgartner et al. [20] proposed a classic road model based on the composition of roads. The road model is divided into three layers, i.e., realistic road network layer, geographic geometric feature layer, and image feature layer. The model shows how the different of road materials and geometric shapes in the real world presented in the images. The model also demonstrates the road features from the perspective of high and low resolutions. More precise information can be extracted from RS image with higher resolution, such as road lanes and zebra crossings. However, higher resolution may introduce more interferences, which will disturb the extraction of global road networks. In the coarse scale, most interferences on road surfaces are eliminated, and prominent road edges are preserved to identify road networks. However, the extracted roads are typically broken and imprecise given the lack of resolution. On this basis, the road extraction methods based on multiscale segmentation are extensively researched
PROPOSED SYSTEM ADVANTAGES:
1.HIGH ACCURACY
2.HIGH EFFICIENCY
SYSTEM REQUIREMENTS
SOFTWARE REQUIREMENTS:
• Programming Language : Python
• Font End Technologies : TKInter/Web(HTML,CSS,JS)
• IDE : Jupyter/Spyder/VS Code
• Operating System : Windows 08/10
HARDWARE REQUIREMENTS:
Processor : Core I3
RAM Capacity : 2 GB
Hard Disk : 250 GB
Monitor : 15″ Color
Mouse : 2 or 3 Button Mouse
Key Board : Windows 08/10
