Search

The land cover classification of very high-resolution (VHR) remote sensing images is a challenging task. VHR images depict many complex objects with various shapes in complicated contexts. The deep learning-based method is a solution for such dif- ficult task and feature extraction. Nevertheless, this method cannot efficiently deal with images with complex scene structures. An improved object-based convolutional neural network (IOCNN) is designed to classify VHR images with zone division and convolutional position sampling techniques in this study. The method can achieve the best performance of each zone at its own optimized scales. Based on multi-scale convolutional deep features extracted from VHR images, the objects with irregular shapes can be classified using the approach. In this study, the zone-level scale adaption and multi-scale recognition of complex objects are achieved. The performance of IOCNN is compared with the state-of-the-art methods for feature extraction, including five object-based CNN approaches and two fully convolutional networks (FCNs). The results show that the classification performance of IOCNN is considerably stronger than that of state-of-the-art methods. The overall accuracies of the land cover classification in IOCNN are 91.65% and 93.49% on two tested images. The results demonstrate the practicability of IOCNN. [ABSTRACT FROM AUTHOR]

Traditional and convolutional neural network (CNN)-based geographic object-based image analysis (GeOBIA) land-cover classification methods prosper in remote sensing and generate numerous distinguished achievements. However, a bottleneck emerges and hinders further improvements in classification results, due to the insufficiency of information provided by very high-spatial resolution images (VHSRIs). To be specific, the phenomenon of different objects with similar spectrum and the lack of topographic information (heights) are natural drawbacks of VHSRIs. Thus, multisource data steps into people's sight and shows a promising future. Firstly, for data fusion, this paper proposed a standard normalized digital surface model (StdnDSM) method which was actually a digital elevation model derived from a digital terrain model (DTM) and digital surface model (DSM) to break through the bottleneck by fusing VHSRI and cloud points. It smoothed and improved the fusion of point cloud and VHSRIs and thus performed well in follow-up classification. The fusion data then were utilized to perform multiresolution segmentation (MRS) and worked as training data for the CNN. Moreover, the grey-level co-occurrence matrix (GLCM) was introduced for a stratified MRS. Secondly, for data processing, the stratified MRS was more efficient than unstratified MRS, and its outcome result was theoretically more rational and explainable than traditional global segmentation. Eventually, classes of segmented polygons were determined by majority voting. Compared to pixel-based and traditional object-based classification methods, majority voting strategy has stronger robustness and avoids misclassifications caused by minor misclassified centre points. Experimental analysis results suggested that the proposed method was promising for object-based classification. [ABSTRACT FROM AUTHOR]

Functional zone reflects city's spatial structures, and as a carrier of social and economic activities, it is of critical significance to urban management, resource allocation and planning. However, most researches on functional zone division are based on a large spatial scale such as blocks or other scales larger than it. Aiming at a subtle fine functional result, the concept of Super Object (SO) was especially explained, also a Super Object - Convolutional Neural Network (SO–CNN) based urban functional zone fine division method with very high resolution (VHR) remote sensing image was proposed. The original image was firstly segmented into different SOs which correspond to the basic functional zone units in geography. A random point generation algorithm was used to generate the voting points for functional zone category identification, and then a trained CNN model was employed to assign functional attributes to those voting points. Then a statistical method was involved to count the frequency of the classified voting points of different functional attributes in each basic functional zone units. By voting process, the functional attribute with the highest frequency was assigned to the basic functional zone unit, which corrected the misclassification results of CNN to some extent. This paper also explored the scale effect of the SO on the final functional zone classification result from two aspects, spatial scale of SO and the sampling window size of CNN model. Because of the natural differences between functional zone division and land cover classification, region based overall accuracy assessment method was used to evaluate functional zone division result. Compared with other methods, SO–CNN method can generate higher accuracy and subtle result, based on which larger spatial scale results can be available by scaling-up, so SO–CNN method plays a great significant role on small scale functional space structure research. • Stressing the concept of Super Object (SO). • SO–CNN combining voting strategy was proposed. • Finer functional zone division result can be obtained. • Exploring optimal scale and image resolution of functional zone division. • Discussing accuracy evaluation methods for functional zone. [ABSTRACT FROM AUTHOR]