CRF learning with CNN features for hyperspectral image segmentation

Alam, Fahim Irfan; Zhou, Jun; Liew, Alan Wee‐Chung; Jia, Xiuping

doi:10.1109/igarss.2016.7730798

Cited by 33 publications

(18 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For dense semantic labeling tasks, post-processing after the deep learning model is a common method to optimize the predictions additionally. The most widely used one is Dense Conditional Random Fields (CRF) [36,37]. As a graph theory-based algorithm, pixel-level labels can be considered as random variables and the relationship between pixels in the image can be considered as edges, these two factors constitute a conditional random field.…”

Section: Dense Conditional Random Fields Based On Superpixelmentioning

confidence: 99%

Dense Semantic Labeling with Atrous Spatial Pyramid Pooling and Decoder for High-Resolution Remote Sensing Imagery

et al. 2018

View full text Add to dashboard Cite

Dense semantic labeling is significant in high-resolution remote sensing imagery research and it has been widely used in land-use analysis and environment protection. With the recent success of fully convolutional networks (FCN), various types of network architectures have largely improved performance. Among them, atrous spatial pyramid pooling (ASPP) and encoder-decoder are two successful ones. The former structure is able to extract multi-scale contextual information and multiple effective field-of-view, while the latter structure can recover the spatial information to obtain sharper object boundaries. In this study, we propose a more efficient fully convolutional network by combining the advantages from both structures. Our model utilizes the deep residual network (ResNet) followed by ASPP as the encoder and combines two scales of high-level features with corresponding low-level features as the decoder at the upsampling stage. We further develop a multi-scale loss function to enhance the learning procedure. In the postprocessing, a novel superpixel-based dense conditional random field is employed to refine the predictions. We evaluate the proposed method on the Potsdam and Vaihingen datasets and the experimental results demonstrate that our method performs better than other machine learning or deep learning methods. Compared with the state-of-the-art DeepLab_v3+ our model gains 0.4% and 0.6% improvements in overall accuracy on these two datasets respectively.

show abstract

Section: Dense Conditional Random Fields Based On Superpixelmentioning

confidence: 99%

Dense Semantic Labeling with Atrous Spatial Pyramid Pooling and Decoder for High-Resolution Remote Sensing Imagery

et al. 2018

View full text Add to dashboard Cite

show abstract

“…where F is a set of random variables {F1, F2,…, FN}; Fi is a pixel vector; X is a set of random variables {x1, x2,…, xN}, where xi is the category label of pixel i; Z(F) is a normalizing factor; and c is a clique in a set of cliques Cg, where g induces a potential φc [23,24]. By calculating Equation (1), the CRF adjusts the category label of each pixel and achieves the goal of correcting the result image.…”

Section: Introductionmentioning

confidence: 99%

An End-to-End and Localized Post-Processing Method for Correcting High-Resolution Remote Sensing Classification Result Images

Pan

Zhao

2020

Remote Sensing

View full text Add to dashboard Cite

Since the result images obtained by deep semantic segmentation neural networks are usually not perfect, especially at object borders, the conditional random field (CRF) method is frequently utilized in the result post-processing stage to obtain the corrected classification result image. The CRF method has achieved many successes in the field of computer vision, but when it is applied to remote sensing images, overcorrection phenomena may occur. This paper proposes an end-to-end and localized post-processing method (ELP) to correct the result images of high-resolution remote sensing image classification methods. ELP has two advantages. (1) End-to-end evaluation: ELP can identify which locations of the result image are highly suspected of having errors without requiring samples. This characteristic allows ELP to be adapted to an end-to-end classification process. (2) Localization: Based on the suspect areas, ELP limits the CRF analysis and update area to a small range and controls the iteration termination condition. This characteristic avoids the overcorrections caused by the global processing of the CRF. In the experiments, ELP is used to correct the classification results obtained by various deep semantic segmentation neural networks. Compared with traditional methods, the proposed method more effectively corrects the classification result and improves classification accuracy.

show abstract

“…In the field of object identification, CRF is a classical segmentation method that can segment rough superpixel classification results into pixel-based classification results [39]. Consider a remote sensing image that contains N pixels and a random field I with random variables {I 1 , I 2 , .…”

Section: Fully Connected Crfmentioning

confidence: 99%

High-Resolution Remote Sensing Image Classification Method Based on Convolutional Neural Network and Restricted Conditional Random Field

Pan

Zhao

2018

Remote Sensing

View full text Add to dashboard Cite

Abstract:Convolutional neural networks (CNNs) can adapt to more complex data, extract deeper characteristics from images, and achieve higher classification accuracy in remote sensing image scene classification and object detection compared to traditional shallow-model methods. However, directly applying common-structure CNNs to pixel-based remote sensing image classification will lead to boundary or outline distortions of the land cover and consumes enormous computation time in the image classification stage. To solve this problem, we propose a high-resolution remote sensing image classification method based on CNN and the restricted conditional random field algorithm (CNN-RCRF). CNN-RCRF adopts CNN superpixel classification instead of pixel-based classification and uses the restricted conditional random field algorithm (RCRF) to refine the superpixel result image into a pixel-based result. The proposed method not only takes advantage of the classification ability of CNNs but can also avoid boundary or outline distortions of the land cover and greatly reduce computation time in classifying images. The effectiveness of the proposed method is tested with two high-resolution remote sensing images, and the experimental results show that the CNN-RCRF outperforms the existing traditional methods in terms of overall accuracy, and CNN-RCRF's computation time is much less than that of traditional pixel-based deep-model methods.

show abstract

CRF learning with CNN features for hyperspectral image segmentation

Cited by 33 publications

References 8 publications

Dense Semantic Labeling with Atrous Spatial Pyramid Pooling and Decoder for High-Resolution Remote Sensing Imagery

Dense Semantic Labeling with Atrous Spatial Pyramid Pooling and Decoder for High-Resolution Remote Sensing Imagery

An End-to-End and Localized Post-Processing Method for Correcting High-Resolution Remote Sensing Classification Result Images

High-Resolution Remote Sensing Image Classification Method Based on Convolutional Neural Network and Restricted Conditional Random Field

Contact Info

Product

Resources

About