The Automatic Extraction of Road Information by the Technology of Object-oriented

Abstract

Road information is rich in urban districts and mining areas. With the increase in spatial resolution and spectral resolution of remote sensing data, it is possible to extract information of narrow roads. However, the traditional manual extraction method using high-spatial-resolution data has shortcomings of low accuracy and low efficiency. Considering the features of SPOT-5 and Gaofen-1 data and the actual road situation of the study areas, the object-oriented method is used in this paper. The main advantage of this method is overcoming the limit of road extraction only using spectral information. A comparison with the results of the traditional supervised classification based on pixels proves that the objectoriented method improves the accuracy by 10% and provides better road information results. In addition, it is a more effective method to extract information for geographical condition monitoring.

Multi-spectral and Topographic Fusion for Automated Road Extraction

“…The proposed method was also able to more satisfactorily remove nonroad pixels than previously possible [16]. More recent OBC method [34,35], with which the proposed method was also benchmarked, yielded less accuracy than the proposed method by approximately 4-5%. It also missed some road pixels while leaving some urban objects misclassified.…”

“…In comparison with other existing works on road extractions, the proposed method has overcome several previous and limitations. They include but not limited to having to rely on user to initialize the estimation [4], depending on threshold that varies from one study area to other [22,23], being unable to removed other urban structures [22,23,34,35] and being suffered from relatively low accuracy [22,23,34,35]. More detailed comparison with the baseline techniques (e.g., those based on single aerial and DSM image), the proposed method has 10% higher accuracy with 0.25 higher Kappa.…”

“…Based on this proposal, several modalities had so far been considered, with more promising results. OBC in SPOT5 and QuickBird yielded about 92-93% accuracy of road extraction, leaving a few buildings and other constructions [34,35]. Similar approach was adopted in WorldView2 images, with 88% accuracy [36].…”

Monitoring Spatial and Temporal Patterns of Rubber Plantation Dynamics Using Time-Series Landsat Images and Google Earth Engine

“…(2) Texture features Texture features refer to the fact that the color and gray level of ground objects change to a certain extent, showing irregular changes locally on the map but showing regular patches at larger scales [35]. However, the texture characteristics of rubber plantations in the leaf-off period are obvious, which can make up for defects of misclassification and misjudgment based solely on spectral information and improve the classification accuracy.…”