Abstract:Nowadays, advanced sensor technology and image processing algorithms allow us to measure different aspects of the objects on the Earth’s surface, from spectral characteristics in optical images, height information in LiDAR data, to spatial information generated by image processing technologies like commercial software eCognition®. However, automatic recognition of objects in remote sensed scenes remains challenging. It is clear that single technology might not be sufficient to obtain reliable classification results (Debes, 2014). Multisensor data, once combined, can contribute to a more comprehensive interpretation of objects on the ground. For example, spectral reflections from optical image cannot recognize objects under shadows, while they can often be easily detected by LiDAR data. On the other hand, LiDAR data alone may fail to discriminate between objects that are quite similar in height.Stacking multi-source data together is a widely applied data fusion technique for classification. These methods first apply feature extraction on each individual data source, after which all feature sources are concatenated into one stacked vector for classification. While such methods are appealing due to their simplicity, they do not always perform better than using a single data source. This is because the value of different components in the stacked feature vector can be significantly unbalanced. As a consequence, theinformation contained by different data sources is not equally represented or measured. Furthermore, the increase in the dimensionality of the stacked features, combined with the limited number of labelled samples, may together lead to the problem of the “curse of dimensionality” (Liao, 2015).Therefore, we present a local graph fusion method to fuse true orthophoto and LiDAR image for urban object recognition. First, object-based spatial and height information are generated on true orthophoto and LiDAR image, respectively. Second, we build a local fusion graph within a sliding window where only the data points with similar spatial and height characteristics are connected. Finally, we solve the problem of multisensor data fusion by projecting multisensor data into a subspace, on which the advantages ofdifferent data sources are well exploited. Experimental results on fusion of true orthophoto and LiDAR image from 'ISPRS Test Project on Urban Classification and 3D Building Reconstruction' demonstrate the potential of the proposed method. Compared to the methods using only single data source or stacking them together, our approach has significant improvements in overall classification accuracy. Both the method’s details and the results of a comprehensive test will be presented at GEOBIA 2016.

Urban Classification by the Fusion of Thermal Infrared Hyperspectral and Visible Data

A Novel Feature Fusion Approach for VHR Remote Sensing Image Classification

Advanced Multi-Sensor Optical Remote Sensing for Urban Land Use and Land Cover Classification: Outcome of the 2018 IEEE GRSS Data Fusion Contest

Relative Total Variation Structure Analysis-Based Fusion Method for Hyperspectral and LiDAR Data Classification

Multi-Scale Feature Extraction and Total Variation Based Fusion Method for HSI and Lidar Data Classification

A New Fusion Approach for Extracting Urban Built-up Areas from Multisource Remotely Sensed Data.

Fusion of optical and liDAR images for urban objects recognition

Processing of Multiresolution Thermal Hyperspectral and Digital Color Data: Outcome of the 2014 IEEE GRSS Data Fusion Contest

Urban Land Use/Land Cover Classification Based on Feature Fusion Fusing Hyperspectral Image and Lidar Data

Fusion of Airborne Discrete-Return LiDAR and Hyperspectral Data for Land Cover Classification

Fusing Ultra-Hyperspectral and High Spatial Resolution Information for Land Cover Classification Based on AISAIBIS Sensor and Phase Camera

LiDAR-Guided Cross-Attention Fusion for Hyperspectral Band Selection and Image Classification

Discriminative Graph-Based Fusion of HSI and LiDAR Data for Urban Area Classification.

Spatial-Spectral-Emissivity Land-Cover Classification Fusing Visible and Thermal Infrared Hyperspectral Imagery

Information fusion of aerial images and LIDAR data in urban areas: vector-stacking, re-classification and post-processing approaches

Open Data for Global Multimodal Land Use Classification: Outcome of the 2017 IEEE GRSS Data Fusion Contest

The Comparison Of Fusion Methods For Hsrrsi Considering The Effectiveness Of Land Cover (Features) Object Recognition Based On Deep Learning

Optimized Spatial Gradient Transfer for Hyperspectral-LiDAR Data Classification

High Resolution Visible Image Completion Of Urban Region Using Corresponding Hyperspectral Image

Semi-Supervised Graph Fusion of Hyperspectral and Lidar Data for Classification