Hyperspectral Image Classification Based on Parameter-Optimized 3D-CNNs Combined with Transfer Learning and Virtual Samples

Li, Xuefeng; Sun, Qiaoqiao; Yue, Meng; Fu, Min; Bourennane, Salah

doi:10.3390/rs10091425

Cited by 33 publications

(15 citation statements)

References 58 publications

(66 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning replaces the hand-crafted feature-engineering process, which requires expert experience and careful parameter settings, with automatic extraction of the meaningful features contained in high-dimensional bands [24]. CNNs have been widely applied to HSI classification tasks [25][26][27][28]. Many studies have successively classified the items in hyperspectral images using 2D-CNNs, which extract features from spatial domains [25,26].…”

Section: Step 1: Hyperpsectral Uav Image Classification For Generatinmentioning

confidence: 99%

“…is the input at position (x + h)(y + w) and (h, w) denotes its offset to (x, y). 3D-CNNs simultaneously extract the spatial and spectral features [27,28]. A 3D-CNN preserves the original input data by avoiding complex data reconstruction and considers the relationships among channels; however, 3D-CNNs are more computationally complex than 2D-CNNs.…”

Section: Step 1: Hyperpsectral Uav Image Classification For Generatinmentioning

confidence: 99%

See 1 more Smart Citation

Discrepancy Analysis for Detecting Candidate Parcels Requiring Update of Land Category in Cadastral Map Using Hyperspectral UAV Images: A Case Study in Jeonju, South Korea

Park

Song

2020

Remote Sensing

View full text Add to dashboard Cite

The non-spatial information of cadastral maps must be repeatedly updated to monitor recent changes in land property and to detect illegal land registrations by tax evaders. Since non-spatial information, such as land category, is usually updated by field-based surveys, it is time-consuming and only a limited area can be updated at a time. Although land categories can be updated by remote sensing techniques, the update is typically performed through manual analysis, namely through a visually interpreted comparison between the newly generated land information and the existing cadastral maps. A cost-effective, fast alternative to the current surveying methods would improve the efficiency of land management. For this purpose, the present study analyzes the discrepancy between the existing cadastral map and the actual land use. Our proposed method operates in two steps. First, an up-to-date land cover map is generated from hyperspectral unmanned aerial vehicle (UAV) images. These images are effectively classified by a hybrid two- and three-dimensional convolutional neural network. Second, a discrepancy map, which contains the ratio of the area that is being used differently from the registered land use in each parcel, is constructed through a three-stage inconsistency comparison. As a case study, the proposed method was evaluated using hyperspectral UAV images acquired at two sites of Jeonju in South Korea. The overall classification accuracies of six land classes at Sites 1 and 2 were 99.93% and 99.75% and those at Sites 1 and 2 are 39.4% and 34.4%, respectively, which had discrepancy ratios of 50% or higher. Finally, discrepancy maps between the land cover maps and existing cadastral maps were generated and visualized. The method automatically reveals the inconsistent parcels requiring updates of their land category. Although the performance of the proposed method depends on the classification results obtained from UAV imagery, the method allows a flexible modification of the matching criteria between the land categories and land coverage. Therefore, it is generalizable to various cadastral systems and the discrepancy ratios will provide practical information and significantly reduce the time and effort for land monitoring and field surveying.

show abstract

Section: Step 1: Hyperpsectral Uav Image Classification For Generatinmentioning

confidence: 99%

Section: Step 1: Hyperpsectral Uav Image Classification For Generatinmentioning

confidence: 99%

Discrepancy Analysis for Detecting Candidate Parcels Requiring Update of Land Category in Cadastral Map Using Hyperspectral UAV Images: A Case Study in Jeonju, South Korea

Park

Song

2020

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The use of transfer learning technique also provides a feasible solution to address the problem of insufficient training samples [46][47][48]. Transfer learning technology is to transfer the knowledge learned from the source model to different but related new tasks, thus reducing both the training time of the new task and the number of labeled samples needed.…”

Section: Introductionmentioning

confidence: 99%

“…Transfer learning technology is to transfer the knowledge learned from the source model to different but related new tasks, thus reducing both the training time of the new task and the number of labeled samples needed. Liu et al suggested an HSI classification method to improve the performance of 3D-CNN model through parameter optimization, transfer learning and virtual samples [47]. By combining the 3-D separable ResNet with cross-sensor transfer learning, an effective approach is presented to classify the HSIs with only a few of labeled samples [48].…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Image Classification Based on Superpixel Pooling Convolutional Neural Network with Transfer Learning

et al. 2021

View full text Add to dashboard Cite

Deep learning-based hyperspectral image (HSI) classification has attracted more and more attention because of its excellent classification ability. Generally, the outstanding performance of these methods mainly depends on a large number of labeled samples. Therefore, it still remains an ongoing challenge how to integrate spatial structure information into these frameworks to classify the HSI with limited training samples. In this study, an effective spectral-spatial HSI classification scheme is proposed based on superpixel pooling convolutional neural network with transfer learning (SP-CNN). The suggested method includes three stages. The first part consists of convolution and pooling operation, which is a down-sampling process to extract the main spectral features of an HSI. The second part is composed of up-sampling and superpixel (homogeneous regions with adaptive shape and size) pooling to explore the spatial structure information of an HSI. Finally, the hyperspectral data with each superpixel as a basic input rather than a pixel are fed to fully connected neural network. In this method, the spectral and spatial information is effectively fused by using superpixel pooling technique. The use of popular transfer learning technology in the proposed classification framework significantly improves the training efficiency of SP-CNN. To evaluate the effectiveness of the SP-CNN, extensive experiments were conducted on three common real HSI datasets acquired from different sensors. With 30 labeled pixels per class, the overall classification accuracy provided by this method on three benchmarks all exceeded 93%, which was at least 4.55% higher than that of several state-of-the-art approaches. Experimental and comparative results prove that the proposed algorithm can effectively classify the HSI with limited training labels.

show abstract

“…The obtained hyperspectral image (HSI) [1][2][3][4], which has a high spectral resolution, is a data cube containing very narrow spectral bands ranging from the visible to infrared spectrum, enabling the fine representation of different land-covers by spectral signatures. Therefore, HSIs have recently been successfully applied in various tasks, such as classification [5][6][7], unmixing [8][9][10], de-noising [11,12] and detection [13][14][15]. However, the high spectral resolution also comes at a cost, i.e., low-spatial-resolution, that is, the acquired real HSI data usually provides coarse spatial information, and thus are incapable of capturing the details of different objects.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Image Super-Resolution Inspired by Deep Laplacian Pyramid Network

He¹,

Liu

2018

Remote Sensing

View full text Add to dashboard Cite

Existing hyperspectral sensors usually produce high-spectral-resolution but low-spatial-resolution images, and super-resolution has yielded impressive results in improving the resolution of the hyperspectral images (HSIs). However, most of the super-resolution methods require multiple observations of the same scene and improve the spatial resolution without fully considering the spectral information. In this paper, we propose an HSI super-resolution method inspired by the deep Laplacian pyramid network (LPN). First, the spatial resolution is enhanced by an LPN, which can exploit the knowledge from natural images without using any auxiliary observations. The LPN progressively reconstructs the high-spatial-resolution images in a coarse-to-fine fashion by using multiple pyramid levels. Second, spectral characteristics between the low- and high-resolution HSIs are studied by the non-negative dictionary learning (NDL), which is proposed to learn the common dictionary with non-negative constraints. The super-resolution results can finally be obtained by multiplying the learned dictionary and its corresponding sparse codes. Experimental results on three hyperspectral datasets demonstrate the feasibility of the proposed method in enhancing the spatial resolution of the HSI with preserving the spectral information simultaneously.

show abstract

Hyperspectral Image Classification Based on Parameter-Optimized 3D-CNNs Combined with Transfer Learning and Virtual Samples

Cited by 33 publications

References 58 publications

Discrepancy Analysis for Detecting Candidate Parcels Requiring Update of Land Category in Cadastral Map Using Hyperspectral UAV Images: A Case Study in Jeonju, South Korea

Discrepancy Analysis for Detecting Candidate Parcels Requiring Update of Land Category in Cadastral Map Using Hyperspectral UAV Images: A Case Study in Jeonju, South Korea

Hyperspectral Image Classification Based on Superpixel Pooling Convolutional Neural Network with Transfer Learning

Hyperspectral Image Super-Resolution Inspired by Deep Laplacian Pyramid Network

Contact Info

Product

Resources

About