Feature Extraction of Hyperspectral Images Based on Deep Boltzmann Machine

Yang, Jiangong; Guo, Yanhui; Wang, Xili

doi:10.1109/lgrs.2019.2937601

Cited by 11 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The radiation conditions of the image assimilate to a medium image D dn . This method is often used as the preprocessing of remote sensing image change detection (Yang et al, 2020).…”

Section: Remote Sensing Image Datamentioning

confidence: 99%

Research on Information Extraction of the Dongting Lake Ecological Wetland Based on Genetic Algorithm Optimized Convolutional Neural Network

Wan

Yin

2022

Front. Ecol. Evol.

View full text Add to dashboard Cite

Dongting Lake is an important lake wetland in China. How to quickly and accurately obtain the basic information of the Dongting Lake ecological wetland is of great + significance for the dynamic monitoring, protection, and sustainable utilization of the wetland. Therefore, this article proposes the information extraction of the Dongting Lake ecological wetland based on genetic algorithm optimized convolutional neural network (GA-CNN), an analysis model combining genetic algorithm (GA) and convolutional neural network (CNN). Firstly, we know the environmental information of Dongting Lake, take Gaofen-1 image as the data source, and use normalized vegetation index and normalized water body index as auxiliary data to preprocess the change detection of remote sensing images to obtain high-precision fitting images. GA-CNN is constructed to efficiently extract the information of the Dongting Lake ecological wetland, and the Relu excitation function is used to improve the phenomenon of gradient disappearance and convergence fluctuation so as to reduce the operation time. Logistic regression is used for feature extraction, and finally the automatic identification and information extraction of the Dongting Lake ecological wetland are realized. The research results show that the method proposed in this article can more deeply dig the information of ground objects, express depth features, and has high accuracy and credibility.

show abstract

“…The radiation conditions of the image assimilate to a medium image D dn . This method is often used as the preprocessing of remote sensing image change detection (Yang et al, 2020).…”

Section: Remote Sensing Image Datamentioning

confidence: 99%

Research on Information Extraction of the Dongting Lake Ecological Wetland Based on Genetic Algorithm Optimized Convolutional Neural Network

Wan

Yin

2022

Front. Ecol. Evol.

View full text Add to dashboard Cite

show abstract

“…DBN is an improved network of restricted Boltzmann machine (RBM), which belongs to unsupervised learning. Reference [ 20 ] introduced local receptive field and weight sharing into Deep Boltzmann Machine (DBM), and established a local-global DBM. However, this method requires more computing resources and increases the corresponding management cost.…”

Section: Related Workmentioning

confidence: 99%

A Method for Extracting Building Information from Remote Sensing Images Based on Deep Learning

Lian-ying

Chen²,

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Semantic segmentation of remote sensing images is an important issue in remote sensing tasks. Existing algorithms can extract information more accurately, but it is difficult to capture the contours of objects and further reveal the interaction information between different objects in the image. Therefore, a deep learning-based method for extracting building information from remote sensing images is proposed. First, the deep learning semantic segmentation model DeepLabv3+ and Mixconv2d are combined, and convolution kernels of different sizes are used for feature recognition. Then, the regularization method based on Rdrop Loss improves the accuracy and efficiency of contour capture for objects of different resolutions, and at the same time improves the consistency of dataset fitting. Finally, the proposed remote sensing image information extraction method is verified based on the self-built dataset. The experimental results show that the proposed algorithm can effectively improve the algorithm efficiency and result accuracy, and has good segmentation performance.

show abstract

“…The deep Boltzmann machine and convolutional neural network (CNN) are combined by Wu, 14 the high-level semantic features of the image were extracted by him, target recognition is achieved, but the underlying information is ignored. The deep Boltzmann machine is optimized by Yang et al, 15 the local-global method was proposed, deep learning methods are combined by him, the amount of training specimens is reduced, the experimental results are superior to the previous feature extraction, but the scheme of memorizing local and global increases the amount of memory at the same time. Aiming at the problem of large specimen volume, an unsupervised feature extraction method based on transfer learning was proposed by Sun and Bourennane, 16 this method has great application prospects, but the method of dimensionality reduction will cause some features to disappear.…”

Section: Image Recognition and Classification Based On Sphere Descripmentioning

confidence: 99%

“…The latest deep learning research has made great progress. The LGDBM method is proposed by Yang et al, 15 the model of deep learning was improved by him, this method achieved 98.02% in target recognition. The DR + WGAN-GP is proposed by Sun and Bourennane, 16 this method is combined with a deep learning model and achieved 98.06% in target recognition.…”

Section: Training and Classificationmentioning

confidence: 99%

Spherical object recognition based on the number of contour edges extracted by fitting and convex hull processing

Zou

Qiu

2020

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

The detection of spherical targets in workpiece shape clustering and fruit classification tasks is challenging. Spherical targets produce low detection accuracy in complex fields, and single-feature processing cannot accurately recognize spheres. Therefore, a novel spherical descriptor (SD) for contour fitting and convex hull processing is proposed. The SD achieves image de-noising by combining flooding processing and morphological operations. The number of polygon-fitted edges is obtained by convex hull processing based on contour extraction and fitting, and two RGB images of the same group of objects are obtained from different directions. The two fitted edges of the same target object obtained at two RGB images are extracted to form a two-dimensional array. The target object is defined as a sphere if the two values of the array are greater than a custom threshold. The first classification result is obtained by an improved K-NN algorithm. Circle detection is then performed on the results using improved Hough circle detection. We abbreviate it as a new Hough transform sphere descriptor (HSD). Experiments demonstrate that recognition of spherical objects is obtained with 98.8% accuracy. Therefore, experimental results show that our method is compared with other latest methods, HSD has higher identification accuracy than other methods.

show abstract

Feature Extraction of Hyperspectral Images Based on Deep Boltzmann Machine

Cited by 11 publications

References 22 publications

Research on Information Extraction of the Dongting Lake Ecological Wetland Based on Genetic Algorithm Optimized Convolutional Neural Network

Research on Information Extraction of the Dongting Lake Ecological Wetland Based on Genetic Algorithm Optimized Convolutional Neural Network

A Method for Extracting Building Information from Remote Sensing Images Based on Deep Learning

Spherical object recognition based on the number of contour edges extracted by fitting and convex hull processing

Contact Info

Product

Resources

About