Unmanned aerial vehicle (UAV) hyperspectral imagery, distinguished by its exceptional spatial granularity and rich spectral diversity, is widely utilized in urban planning, vegetation monitoring. UAV hyperspectral ima...
详细信息
ISBN:
(纸本)9798350379860;9798350379877
Unmanned aerial vehicle (UAV) hyperspectral imagery, distinguished by its exceptional spatial granularity and rich spectral diversity, is widely utilized in urban planning, vegetation monitoring. UAV hyperspectral images classification is a crucial application for facilitating feature monitoring. However, the complex textures and low signal-to-noise ratio inherent in UAV hyperspectral images make their classification a formidable challenge. Hence, in this article, an innovative classification method that integrates a Graph Convolutional Network (GCN) with Linear Discriminant Analysis-Felzenszwalb (LDA-Felzenszwalb) superpixelsegmentation method was proposed. Firstly, the UAV hyperspectral images are clustered based on the LDA-Felzenszwalb algorithm, which employs downscaling and spectral-spatial similarity. Then, the encoded superpixel images are subjected to feature extraction via graph convolutional networks, aiming to uncover the latent spatial topological relationships within the data. Ultimately, the UAV hyperspectral images are classified with high precision. The effectiveness of the proposed method is demonstrated using the WHU-HI-HongHu dataset, where it achieves an overall classification accuracy of 92.41%, outperforming comparison methods by at least 11.38%. The results show the classification method is highly effective when applied to high spatial resolution hyperspectral images.
Visual based route and boundary detection is a key technology in agricultural automatic navigation systems. The variable illumination and lack of training samples has a bad effect on visual route detection in unstruct...
详细信息
Visual based route and boundary detection is a key technology in agricultural automatic navigation systems. The variable illumination and lack of training samples has a bad effect on visual route detection in unstructured farmland environments. In order to improve the robustness of the boundary detection under different illumination conditions, an image segmentationalgorithm based on support vector machine was proposed. A superpixel segmentation algorithm was adopted to solve the lack of training samples for a support vector machine. A sufficient number of superpixel samples were selected for extraction of color and texture features, thus a 19-dimensional feature vector was formed. Then, the support vector machine model was trained and used to identify the paddy ridge field in the new picture. The recognition F1 score can reach 90.7%. Finally, Hough transform detection was used to extract the boundary of the ridge field. The total running time of the proposed algorithm is within 0.8 s and can meet the real-time requirements of agricultural machinery.
With the development of computer technology and artificial intelligence technology,the authenticity of images has been seriously ***,how to determine the authenticity of images has become an important research directi...
详细信息
With the development of computer technology and artificial intelligence technology,the authenticity of images has been seriously ***,how to determine the authenticity of images has become an important research direction in the field of digital *** present,image splicing is one of the most common methods of image *** sources of images have different noise levels,and this paper proposes a method to locate the image splicing tamper region by detecting the inconsistency of the local noise level of the ***,the image to be detected is segmented into pixel blocks with similar features by using SLIC superpixelsegmentation ***,the noise level estimation method based on principal component analysis is used to calculate the local noise level of each image ***,using three clustering algorithms to cluster the results of the estimated noise level,and the splicing tamper region of the image is located according to the clustering *** experiment results show that the proposed method in this paper can effectively locate the splicing tamper region and retain more edge information of the detected region.
暂无评论