scholarly journals BUILDING SHADOW DETECTION FROM GHOST IMAGERY

2018 ◽  
Vol XLII-3 ◽  
pp. 2527-2530
Author(s):  
G. Zhou ◽  
J. Sha ◽  
T. Yue ◽  
Q. Wang ◽  
X. Liu ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Zenith Angle ◽  
Large Scale ◽  
Region Growing ◽  
Remote Sensing Image ◽  
Difficult Problem ◽  
Shadow Detection ◽  
Ghost Image ◽  
Basic Features ◽  
Remote Sensing Image Processing

Shadow is one of the basic features of remote sensing image, it expresses a lot of information of the object which is loss or interference, and the removal of shadow is always a difficult problem to remote sensing image processing. In this paper, it is mainly analyzes the characteristics and properties of shadows from the ghost image (traditional orthorectification). The DBM and the interior and exterior orientation elements of the image are used to calculate the zenith angle of sun. Then this paper combines the scope of the architectural shadows which has be determined by the zenith angle of sun with the region growing method to make the detection of architectural shadow areas. This method lays a solid foundation for the shadow of the repair from the ghost image later. It will greatly improve the accuracy of shadow detection from buildings and make it more conducive to solve the problem of urban large-scale aerial imagines.

High-Throughput, Resource-Efficient Multi-Dimensional Parallel Architecture for Space-Borne Sea-Land Segmentation

2020 ◽  
pp. 2150027
Author(s):  
Cunguang Zhang ◽  
Hongxun Jiang ◽  
Riwei Pan ◽  
Haiheng Cao ◽  
Mingliang Zhou
Keyword(s):  
Remote Sensing ◽  
High Throughput ◽  
Large Scale ◽  
Parallel Architecture ◽  
Remote Sensing Image ◽  
Satellite System ◽  
Peak Performance ◽  
Remote Sensing Images ◽  
Block Level ◽  
Remote Sensing Image Processing

Sea-land segmentation based on edge detection is commonly utilized in ship detection, coastline extraction, and satellite system applications due to its high accuracy and rapid speed. Pixel-level distribution statistics do not currently satisfy the requirements for high-resolution, large-scale remote sensing image processing. To address the above problem, in this paper, we propose a high-throughput hardware architecture for sea-land segmentation based on multi-dimensional parallel characteristics. The proposed architecture is well suited to wide remote sensing images. Efficient multi-dimensional block level statistics allow for relatively infrequent pixel-level memory access; a boundary block tracking process replaces the whole-image scanning process, markedly enhancing efficiency. The tracking efficiency is further improved by a convenient two-step scanning strategy that feeds back the path state in a timely manner for a large number of blocks in the same direction appearing in the algorithm. The proposed architecture was deployed on Xilinx Virtex k7-410t to find that its practical processing time for a [Formula: see text] remote sensing image is only about 0.4[Formula: see text]s. The peak performance is 1.625[Formula: see text]gbps, which is higher than other FPGA implementations of segmentation algorithms. The proposed structure is highly competitive in processing wide remote sensing images.

An Object-Oriented Shadow Detection Approach of Remote Sensing Image

2011 ◽  
Vol 474-476 ◽  
pp. 1038-1043 ◽  
Author(s):  
Xiao Le Shen ◽  
Zhen Feng Shao ◽  
Hui Luo ◽  
Wei Cheng
Keyword(s):  
Remote Sensing ◽  
Image Interpretation ◽  
Object Oriented ◽  
Remote Sensing Image ◽  
Shadow Detection ◽  
Remote Sensing Images ◽  
Region Detection ◽  
Detection Approach ◽  
Object Oriented Approach ◽  
Remote Sensing Image Processing

Shadows widely exist in high-resolution remote sensing images and affect image interpretation in certain degree. Improving the accuracy and efficiency of shadow region detection is always a significant problem in remote sensing image processing field. In this paper, an object-oriented shadow detection approach of remote sensing image is proposed on the basis of analyzing the characteristics of the shadow object. Experimental results indicate the efficiency and validity of our object-oriented approach for shadow detection compared with conventional pixel-level methods.

MILL: Channel Attention–based Deep Multiple Instance Learning for Landslide Recognition

2021 ◽  
Vol 17 (2s) ◽  
pp. 1-11
Author(s):  
Xiaochuan Tang ◽  
Mingzhe Liu ◽  
Hao Zhong ◽  
Yuanzhen Ju ◽  
Weile Li ◽  
...  
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Large Scale ◽  
Remote Sensing Image ◽  
Disaster Risk ◽  
Multiple Instance Learning ◽  
Remote Sensing Images ◽  
Loess Area ◽  
Remote Sensing Image Classification ◽  
Natural Disaster Risk

Landslide recognition is widely used in natural disaster risk management. Traditional landslide recognition is mainly conducted by geologists, which is accurate but inefficient. This article introduces multiple instance learning (MIL) to perform automatic landslide recognition. An end-to-end deep convolutional neural network is proposed, referred to as Multiple Instance Learning–based Landslide classification (MILL). First, MILL uses a large-scale remote sensing image classification dataset to build pre-train networks for landslide feature extraction. Second, MILL extracts instances and assign instance labels without pixel-level annotations. Third, MILL uses a new channel attention–based MIL pooling function to map instance-level labels to bag-level label. We apply MIL to detect landslides in a loess area. Experimental results demonstrate that MILL is effective in identifying landslides in remote sensing images.

Sign in / Sign up

Export Citation Format

Share Document