Rapid data annotation for sand-like granular instance segmentation using mask-RCNN

Zhang, Zhiyong; Yin, Xiaolei; Yan, Zhiyuan

doi:10.1016/j.autcon.2021.103994

Cited by 19 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparing the four methods, figure 6(b) shows that YOLACT [22][23][24] cannot handle sticky and complex images with too much noise well, and a considerable amount of redundant recognition and undersegmentation occur because of its focus on speed improvement at the expense of overall accuracy. Figure 6(c) shows that Mask R-CNN [25][26][27], as a classic instance segmentation approach, does not achieve good results in segmenting coal dust adhesive images because the influence of particle agglomeration makes the accurate identification of particle outlines difficult, which results in mis-segmentation. Figure 6(d) shows that BlendMask [28][29][30] has slightly improved the accuracy and speed compared with Mask R-CNN, whereas if the segmented target is irregular, the segmentation accuracy is significantly reduced, resulting in more improper segmentation.…”

Section: Coal Dust Image Segmentationmentioning

confidence: 99%

Effective estimation model of coal dust characterization parameters with image sensing

Wang,

Liu,

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Coal dust leakage occurs most often during transportation to a power plant. Owing to the transportation method, the transported high-pressure coal dust can damage weak points in the pipeline wall and leak into the air, leading to serious safety hazards. To address this, this study proposes a coal dust parameter estimation model that combines domain-adaptive segmentation with extreme corrosion and a particle mass prediction regression model to extract the key characteristic signals of leaking coal dust and evaluate the production environment safety status. First, the connected domain is applied to segment the overlapping particles and extract two-dimensional image information. Subsequently, a regression model was constructed to predict the particle mass, which was mapped with the coal dust thickness model, density, and projected area and applied to environmental dust concentration characterization. The experiment samples included 3000 coal dust images captured from production links in power plants. A statistical analysis showed that the proposed model improved the accuracy and reliability of coal dust detection.

show abstract

Section: Coal Dust Image Segmentationmentioning

confidence: 99%

Effective estimation model of coal dust characterization parameters with image sensing

Wang,

Liu,

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Figure 2 illustrates the architecture of Mask-RCNN. At a high level, Mask-RCNN consists of several components: Backbone, Region Proposal Network (RPN), ROI Classifier and Bounding Box Regressor, and Segmentation Masks [20]. The backbone is a Feature Pyramid Network (FPN) [21][22] style feature extractor.…”

Section: Mask-rcnn Architecturementioning

confidence: 99%

Multiclass Segmentation of Pulmonary Diseases using Convolutional Neural Network

et al. 2022

View full text Add to dashboard Cite

Pulmonary disease has affected tens of millions of people in the world. This disease has also become the cause of death of millions of its sufferers every year. In addition, lung disease has also become the cause of other respiratory complications, which also causes the death of the sufferer. The diagnosis of pulmonary diseases through medical imaging is a significant challenge in computer vision and medical image processing. The difficulty is due to the wide variety in infected areas' shape, dimension, and location. Another challenge is to differentiate one lung disease from the other. Discriminating pulmonary diseases is a notable concern in the diagnosis of pulmonary disease. We have adopted the deep learning convolutional neural network in this study to address these challenges. Seven models were constructed using the Mask Region-based Convolutional Neural Network (Mask-RCNN) architecture to detect and segment infected areas within the lung region from CT scan imagery. The evaluation results show that the best model obtained scores of 91.98%, 85.25%, and 93.75% for DSC, MIoU, and mAP, respectively. The segmentation results are then visualized.

show abstract

“…The precision and recall rates were 97.31% and 95.70%, respectively. Zhang et al (2022) developed a mask-labeling methodology for particles with a varying degree of overlap that can establish a large and diverse training set without manual labeling. This could be an efficient sample-labeling method.…”

Section: Introductionmentioning

confidence: 99%

Overlapped tobacco shred image segmentation and area computation using an improved Mask RCNN network and COT algorithm

Wang

Jia

et al. 2023

Front. Plant Sci.

View full text Add to dashboard Cite

IntroductionThe classification of the four tobacco shred varieties, tobacco silk, cut stem, expanded tobacco silk, and reconstituted tobacco shred, and the subsequent determination of tobacco shred components, are the primary tasks involved in calculating the tobacco shred blending ratio. The identification accuracy and subsequent component area calculation error directly affect the composition determination and quality of the tobacco shred. However, tiny tobacco shreds have complex physical and morphological characteristics; in particular, there is substantial similarity between the expanded tobacco silk and tobacco silk varieties, and this complicates their classification. There must be a certain amount of overlap and stacking in the distribution of tobacco shreds on the actual tobacco quality inspection line. There are 24 types of overlap alone, not to mention the stacking phenomenon. Self-winding does not make it easier to distinguish such varieties from the overlapped types, posing significant difficulties for machine vision-based tobacco shred classification and component area calculation tasks.MethodsThis study focuses on two significant challenges associated with identifying various types of overlapping tobacco shreds and acquiring overlapping regions to calculate overlapping areas. It develops a new segmentation model for tobacco shred images based on an improved Mask region-based convolutional neural network (RCNN). Mask RCNN is used as the segmentation network’s mainframe. Convolutional network and feature pyramid network (FPN) in the backbone are replaced with Densenet121 and U-FPN, respectively. The size and aspect ratios of anchors parameters in region proposal network (RPN) are optimized. An algorithm for the area calculation of the overlapped tobacco shred region (COT) is also proposed, which is applied to overlapped tobacco shred mask images to obtain overlapped regions and calculate the overlapped area.ResultsThe experimental results showed that the final segmentation accuracy and recall rates are 89.1% and 73.2%, respectively. The average area detection rate of 24 overlapped tobacco shred samples increases from 81.2% to 90%, achieving high segmentation accuracy and overlapped area calculation accuracy.DiscussionThis study provides a new implementation method for the type identification and component area calculation of overlapped tobacco shreds and a new approach for other similar overlapped image segmentation tasks.

show abstract

Rapid data annotation for sand-like granular instance segmentation using mask-RCNN

Cited by 19 publications

References 13 publications

Effective estimation model of coal dust characterization parameters with image sensing

Effective estimation model of coal dust characterization parameters with image sensing

Multiclass Segmentation of Pulmonary Diseases using Convolutional Neural Network

Overlapped tobacco shred image segmentation and area computation using an improved Mask RCNN network and COT algorithm

Contact Info

Product

Resources

About