Large Kernel Separable Mixed ConvNet for Remote Sensing Scene Classification
Open Access
- 15 January 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 4294-4303
Abstract
Among tasks related to intelligent interpretation of remote sensing data, scene classification mainly focuses on the holistic information of the entire scene. Compared to pixel-level or object-based tasks, it involves a richer semantic context, making it more challenging. With the rapid advancement of deep learning, convolutional neural networks (CNNs) have found widespread applications across various domains, and some work has introduced them into scene classification tasks. However, traditional convolution operations involve sliding small convolutional kernels across an image, primarily focusing on local details within a small receptive field. To achieve better modeling of the entire image, the smaller receptive field limits the ability of convolution operation to capture features over a broader range. To this end, we introduce a large kernel CNNs into the scene classification task to expand the receptive field of the mode, which allows us to capture comprehensive non-local information while still acquiring rich local details. However, in addition to encoding spatial association, the effective information within the feature maps is also strongly channel-related. Therefore, to fully model this channel dependency, a novel channel separation and mixing module has been designed to realize feature correlation in the channel dimension. The combination of them forms a Large Kernel Separable Mixed Convnet (LSMNet), enabling the model to capture effective dependencies of feature maps in both spatial and channel dimensions, thus achieving enhanced feature expression. Extensive experiments conducted on three datasets have also validated the effectiveness of the proposed method.Keywords
Funding Information
- National Key R&D Program Projects (2016YFC08033103)
- National Natural Science Foundation of China (41001226)
- China Postdoctoral Science Foundation (2015M582831)
- Special Funding for Basic Scientific Research Business Expenses of Henan Universities (NSFRF180329)
This publication has 68 references indexed in Scilit:
- Leveraging mayfly optimization with deep learning for secure remote sensing scene image classificationComputers and Electrical Engineering, 2023
- PVT v2: Improved baselines with pyramid vision transformerComputational Visual Media, 2022
- DeepScene: Scene classification via convolutional neural network with spatial pyramid poolingExpert Systems with Applications, 2022
- Urban scene understanding based on semantic and socioeconomic features: From high-resolution remote sensing imagery to multi-source geographic datasetsISPRS Journal of Photogrammetry and Remote Sensing, 2021
- Exploring semantic elements for urban scene recognition: Deep integration of high-resolution imagery and OpenStreetMap (OSM)ISPRS Journal of Photogrammetry and Remote Sensing, 2019
- Sigmoid-weighted linear units for neural network function approximation in reinforcement learningNeural Networks, 2018
- A scene image classification technique for a ubiquitous visual surveillance systemMultimedia Tools and Applications, 2018
- Binary patterns encoded convolutional neural networks for texture recognition and remote sensing scene classificationISPRS Journal of Photogrammetry and Remote Sensing, 2018
- Multi-class geospatial object detection and geographic image classification based on collection of part detectorsISPRS Journal of Photogrammetry and Remote Sensing, 2014
- New color GPHOG descriptors for object and scene image classificationMachine Vision and Applications, 2013