Initial Study on Multi-Scale Patch-Based Classification of Paper Fibers Based on EfficientNet Using Consumer Digital Camera

Urban object segmentation and classification tasks are critical data processing steps in scene understanding, intelligent vehicles and 3D high-precision maps. Semantic segmentation of 3D point clouds is the foundational step in object recognition. To identify the intersecting objects and improve the accuracy of classification, this paper proposes a segment-based classification method for 3D point clouds. This method firstly divides points into multi-scale supervoxels and groups them by proposed inverse node graph (IN-Graph) construction, which does not need to define prior information about the node, it divides supervoxels by judging the connection state of edges between them. This method reaches minimum global energy by graph cutting, obtains the structural segments as completely as possible, and retains boundaries at the same time. Then, the random forest classifier is utilized for supervised classification. To deal with the mislabeling of scattered fragments, higher-order CRF with small-label cluster optimization is proposed to refine the classification results. Experiments were carried out on mobile laser scan (MLS) point dataset and terrestrial laser scan (TLS) points dataset, and the results show that overall accuracies of 97.57% and 96.39% were obtained in the two datasets. The boundaries of objects were retained well, and the method achieved a good result in the classification of cars and motorcycles. More experimental analyses have verified the advantages of the proposed method and proved the practicability and versatility of the method.

Download Full-text

Combining rotation forest and multi-scale segmentation for the classification of hyperspectral data

2015 7th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS) ◽

10.1109/whispers.2015.8075390 ◽

2015 ◽

Author(s):

Jike Chen ◽

Junshi Xia ◽

Peijun Du ◽

Jocelyn Chanussot

Keyword(s):

Hyperspectral Data ◽

Rotation Forest ◽

Multi Scale

Download Full-text

Segmentation and Multi-Scale Convolutional Neural Network-Based Classification of Airborne Laser Scanner Data

Sensors ◽

10.3390/s18103347 ◽

2018 ◽

Vol 18 (10) ◽

pp. 3347 ◽

Cited By ~ 14

Author(s):

Zhishuang Yang ◽

Bo Tan ◽

Huikun Pei ◽

Wanshou Jiang

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Laser Scanning ◽

Laser Scanner ◽

Region Growing ◽

Point Clouds ◽

Classification Method ◽

Multi Scale ◽

Airborne Laser

The classification of point clouds is a basic task in airborne laser scanning (ALS) point cloud processing. It is quite a challenge when facing complex observed scenes and irregular point distributions. In order to reduce the computational burden of the point-based classification method and improve the classification accuracy, we present a segmentation and multi-scale convolutional neural network-based classification method. Firstly, a three-step region-growing segmentation method was proposed to reduce both under-segmentation and over-segmentation. Then, a feature image generation method was used to transform the 3D neighborhood features of a point into a 2D image. Finally, feature images were treated as the input of a multi-scale convolutional neural network for training and testing tasks. In order to obtain performance comparisons with existing approaches, we evaluated our framework using the International Society for Photogrammetry and Remote Sensing Working Groups II/4 (ISPRS WG II/4) 3D labeling benchmark tests. The experiment result, which achieved 84.9% overall accuracy and 69.2% of average F1 scores, has a satisfactory performance over all participating approaches analyzed.

Download Full-text

Classification of Pulmonary Emphysema in CT Images Based on Multi-Scale Deep Convolutional Neural Networks

2018 25th IEEE International Conference on Image Processing (ICIP) ◽

10.1109/icip.2018.8451514 ◽

2018 ◽

Cited By ~ 2

Author(s):

Liying Peng ◽

Lanfen Lin ◽

Hongjie Hu ◽

Huali Li ◽

Xiaoli Ling ◽

...

Keyword(s):

Neural Networks ◽

Convolutional Neural Networks ◽

Pulmonary Emphysema ◽

Ct Images ◽

Deep Convolutional Neural Networks ◽

Multi Scale

Download Full-text

Multi-scale superpixel spectral–spatial classification of hyperspectral images

International Journal of Remote Sensing ◽

10.1080/01431161.2016.1225175 ◽

2016 ◽

Vol 37 (20) ◽

pp. 4905-4922 ◽

Cited By ~ 20

Author(s):

Shanshan Li ◽

Li Ni ◽

Xiuping Jia ◽

Lianru Gao ◽

Bing Zhang ◽

...

Keyword(s):

Hyperspectral Images ◽

Spatial Classification ◽

Multi Scale

Download Full-text

Pine wilt disease detection in high-resolution UAV images using object-oriented classification

Journal of Forestry Research ◽

10.1007/s11676-021-01420-x ◽

2021 ◽

Author(s):

Zhao Sun ◽

Yifu Wang ◽

Lei Pan ◽

Yunhong Xie ◽

Bo Zhang ◽

...

Keyword(s):

High Resolution ◽

Large Scale ◽

Digital Camera ◽

Object Oriented ◽

Pine Wilt Disease ◽

Wilt Disease ◽

Segmentation Algorithm ◽

Tree Crown ◽

Multi Scale ◽

Pine Wilt

AbstractPine wilt disease (PWD) is currently one of the main causes of large-scale forest destruction. To control the spread of PWD, it is essential to detect affected pine trees quickly. This study investigated the feasibility of using the object-oriented multi-scale segmentation algorithm to identify trees discolored by PWD. We used an unmanned aerial vehicle (UAV) platform equipped with an RGB digital camera to obtain high spatial resolution images, and multi-scale segmentation was applied to delineate the tree crown, coupling the use of object-oriented classification to classify trees discolored by PWD. Then, the optimal segmentation scale was implemented using the estimation of scale parameter (ESP2) plug-in. The feature space of the segmentation results was optimized, and appropriate features were selected for classification. The results showed that the optimal scale, shape, and compactness values of the tree crown segmentation algorithm were 56, 0.5, and 0.8, respectively. The producer’s accuracy (PA), user’s accuracy (UA), and F1 score were 0.722, 0.605, and 0.658, respectively. There were no significant classification errors in the final classification results, and the low accuracy was attributed to the low number of objects count caused by incorrect segmentation. The multi-scale segmentation and object-oriented classification method could accurately identify trees discolored by PWD with a straightforward and rapid processing. This study provides a technical method for monitoring the occurrence of PWD and identifying the discolored trees of disease using UAV-based high-resolution images.

Download Full-text