Semantic segmentation based on DeepLabV3+ and superpixel optimization

Feng-lei, 任凤雷 Ren; Xin, 何昕 He; Zhong-hui, 魏仲慧 Wei; You, 吕 游 L; Mu-yu, 李沐雨 Li

doi:10.3788/ope.20192712.2722

Cited by 7 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This article used the original network architecture developed from FCN and VGG16. (7) DeepLabv3+ [68] is considered one of the most advanced algorithms for semantic segmentation. It uses the encoding-decoding structure for multi-scale information fusion while retaining the original dilated convolution and ASSP layer.…”

Section: Comparison With Other Methodsmentioning

confidence: 99%

A Novel Squeeze-and-Excitation W-Net for 2D and 3D Building Change Detection with Multi-Source and Multi-Feature Remote Sensing Data

Zhang

Wang

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Building Change Detection (BCD) is one of the core issues in earth observation and has received extensive attention in recent years. With the rapid development of earth observation technology, the data source of remote sensing change detection is continuously enriched, which provides the possibility to describe the spatial details of the ground objects more finely and to characterize the ground objects with multiple perspectives and levels. However, due to the different physical mechanisms of multi-source remote sensing data, BCD based on heterogeneous data is a challenge. Previous studies mostly focused on the BCD of homogeneous remote sensing data, while the use of multi-source remote sensing data and considering multiple features to conduct 2D and 3D BCD research is sporadic. In this article, we propose a novel and general squeeze-and-excitation W-Net, which is developed from U-Net and SE-Net. Its unique advantage is that it can not only be used for BCD of homogeneous and heterogeneous remote sensing data respectively but also can input both homogeneous and heterogeneous remote sensing data for 2D or 3D BCD by relying on its bidirectional symmetric end-to-end network architecture. Moreover, from a unique perspective, we use image features that are stable in performance and less affected by radiation differences and temporal changes. We innovatively introduced the squeeze-and-excitation module to explicitly model the interdependence between feature channels so that the response between the feature channels is adaptively recalibrated to improve the information mining ability and detection accuracy of the model. As far as we know, this is the first proposed network architecture that can simultaneously use multi-source and multi-feature remote sensing data for 2D and 3D BCD. The experimental results in two 2D data sets and two challenging 3D data sets demonstrate that the promising performances of the squeeze-and-excitation W-Net outperform several traditional and state-of-the-art approaches. Moreover, both visual and quantitative analyses of the experimental results demonstrate competitive performance in the proposed network. This demonstrates that the proposed network and method are practical, physically justified, and have great potential application value in large-scale 2D and 3D BCD and qualitative and quantitative research.

show abstract

Section: Comparison With Other Methodsmentioning

confidence: 99%

A Novel Squeeze-and-Excitation W-Net for 2D and 3D Building Change Detection with Multi-Source and Multi-Feature Remote Sensing Data

Zhang

Wang

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…At the beginning of this experiment, the image is similar to pixel merging processing, that is, using the super pixel method (Ren et al, 2019), SLIC super pixel segmentation algorithm is introduced in the prediction, and the input image is divided into a super pixel image. The traditional DeeplabV3+ output operation of these super pixel image regions is used to obtain an accurate segmentation image.…”

Section: Forecast Strategy Validationmentioning

confidence: 99%

A method of cotton root segmentation based on edge devices

Tang

Zhu

et al. 2023

Front. Plant Sci.

View full text Add to dashboard Cite

The root is an important organ for plants to absorb water and nutrients. In situ root research method is an intuitive method to explore root phenotype and its change dynamics. At present, in situ root research, roots can be accurately extracted from in situ root images, but there are still problems such as low analysis efficiency, high acquisition cost, and difficult deployment of image acquisition devices outdoors. Therefore, this study designed a precise extraction method of in situ roots based on semantic segmentation model and edge device deployment. It initially proposes two data expansion methods, pixel by pixel and equal proportion, expand 100 original images to 1600 and 53193 respectively. It then presents an improved DeeplabV3+ root segmentation model based on CBAM and ASPP in series is designed, and the segmentation accuracy is 93.01%. The root phenotype parameters were verified through the Rhizo Vision Explorers platform, and the root length error was 0.669%, and the root diameter error was 1.003%. It afterwards designs a time-saving Fast prediction strategy. Compared with the Normal prediction strategy, the time consumption is reduced by 22.71% on GPU and 36.85% in raspberry pie. It ultimately deploys the model to Raspberry Pie, realizing the low-cost and portable root image acquisition and segmentation, which is conducive to outdoor deployment. In addition, the cost accounting is only $247. It takes 8 hours to perform image acquisition and segmentation tasks, and the power consumption is as low as 0.051kWh. In conclusion, the method proposed in this study has good performance in model accuracy, economic cost, energy consumption, etc. This paper realizes low-cost and high-precision segmentation of in-situ root based on edge equipment, which provides new insights for high-throughput field research and application of in-situ root.

show abstract

“…Among them, the most common lightweight residual connectivity networks include the MobileNet family. Compared with MobileNetv1 [25], the biggest advantage of MobileNetv2 [26] is the use of residual connectivity to prevent the network from degradation. Therefore, in this paper, MobileNetv2, which has a stronger network feature extraction capability, is selected as the backbone feature extraction network of Mobile-Deep.…”

Section: Mobilenetv2 Structure Parameter Adjustmentmentioning

confidence: 99%

Mobile-Deep Based PCB Image Segmentation Algorithm Research

Liu,

Ke,

Lin

2023

CMC

View full text Add to dashboard Cite

Aiming at the problems of inaccurate edge segmentation, the hole phenomenon of segmenting large-scale targets, and the slow segmentation speed of printed circuit boards (PCB) in the image segmentation process, a PCB image segmentation model Mobile-Deep based on DeepLabv3+ semantic segmentation framework is proposed. Firstly, the DeepLabv3+ feature extraction network is replaced by the lightweight model MobileNetv2, which effectively reduces the number of model parameters; secondly, for the problem of positive and negative sample imbalance, a new loss function is composed of Focal Loss combined with Dice Loss to solve the category imbalance and improve the model discriminative ability; in addition, a more efficient atrous spatial pyramid pooling (E-ASPP) module is proposed. In addition, a more efficient E-ASPP module is proposed, and the Roberts crossover operator is chosen to sharpen the image edges to improve the model accuracy; finally, the network structure is redesigned to further improve the model accuracy by drawing on the multi-scale feature fusion approach. The experimental results show that the proposed segmentation algorithm achieves an average intersection ratio of 93.45%, a precision of 94.87%, a recall of 93.65%, and a balance score of 93.64% on the PCB test set, which is more accurate than the common segmentation algorithms Hrnetv2, UNet, PSPNet, and PCBSegClassNet, and the segmentation speed is faster.

show abstract

Semantic segmentation based on DeepLabV3+ and superpixel optimization

Cited by 7 publications

References 0 publications

A Novel Squeeze-and-Excitation W-Net for 2D and 3D Building Change Detection with Multi-Source and Multi-Feature Remote Sensing Data

A Novel Squeeze-and-Excitation W-Net for 2D and 3D Building Change Detection with Multi-Source and Multi-Feature Remote Sensing Data

A method of cotton root segmentation based on edge devices

Mobile-Deep Based PCB Image Segmentation Algorithm Research

Contact Info

Product

Resources

About