Weakly supervised learning with convolutional neural networks for power line localization

Lee, Sang Jun; Yun, Jong Pil; Choi, Hyeyeon; Kwon, Wookyong; Koo, Gyogwon; Kim, Sang Woo

doi:10.1109/ssci.2017.8285410

Cited by 20 publications

(21 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where R i , P i , T P i , F N i , F P i are pixel-level recall, pixellevel precision, number of true positive pixels, number of false negative pixels, and number of false positive pixels of the ith image, respectively, and N is the number of test images. First, we compare our proposed LS-Net with the weakly supervised learning with CNNs (WSL-CNN) approach proposed in [20] on the publicly available ground truth of power line dataset (Infrared-IR and Visible Light-VL) [43], which is one of the most widely used power line datasets. The LS-Net and the WSL-CNN approaches share a similar objective that is to localize power lines by using cheaper ground-truth data (GTD) than pixel-level GTD (e.g., image-level class information and line end-point information).…”

Section: Comparisons With the State-of-the-art Resultsmentioning

confidence: 99%

“…The LS-Net and the WSL-CNN approaches share a similar objective that is to localize power lines by using cheaper ground-truth data (GTD) than pixel-level GTD (e.g., image-level class information and line end-point information). For a fair comparison, we convert line segment maps generated by the LS-Net to pixel-level segmentation maps using a similar procedure as applied in [20]. First, the pixel-level segmentation maps, S, are generated as follows:…”

Section: Comparisons With the State-of-the-art Resultsmentioning

confidence: 99%

“…Although treating wire detection as a semantic segmentation task has been proved to be a powerful approach for detecting wires [25], its requirement of pixel-level annotated ground-truth data makes it less practical than traditional computer vision approaches. Sang et al proposed to use weekly supervised learning with CNNs for localizing power lines in pixel-level precision by only using image-level class information [20]. First, a classifier adapted from the VGG19 is applied to classify subregions (128 × 128) from an input image (512×512) by using a sliding window approach.…”

Section: Dl-based Methodsmentioning

confidence: 99%

See 2 more Smart Citations

LS-Net: fast single-shot line-segment detector

Nguyen

Jenssen

Roverso³

2020

Machine Vision and Applications

View full text Add to dashboard Cite

In unmanned aerial vehicle (UAV) flights, power lines are considered as one of the most threatening hazards and one of the most difficult obstacles to avoid. In recent years, many vision-based techniques have been proposed to detect power lines to facilitate self-driving UAVs and automatic obstacle avoidance. However, most of the proposed methods are typically based on a common three-step approach: (i) edge detection, (ii) the Hough transform, and (iii) spurious line elimination based on power line constrains. These approaches not only are slow and inaccurate but also require a huge amount of effort in post-processing to distinguish between power lines and spurious lines. In this paper, we introduce LS-Net, a fast single-shot line-segment detector, and apply it to power line detection. The LS-Net is by design fully convolutional, and it consists of three modules: (i) a fully convolutional feature extractor, (ii) a classifier, and (iii) a line segment regressor. Due to the unavailability of large datasets with annotations of power lines, we render synthetic images of power lines using the physically based rendering approach and propose a series of effective data augmentation techniques to generate more training data. With a customized version of the VGG-16 network as the backbone, the proposed approach outperforms existing state-of-the-art approaches. In addition, the LS-Net can detect power lines in near real time. This suggests that our proposed approach has a promising role in automatic obstacle avoidance and as a valuable component of self-driving UAVs, especially for automatic autonomous power line inspection.

show abstract

Section: Comparisons With the State-of-the-art Resultsmentioning

confidence: 99%

Section: Comparisons With the State-of-the-art Resultsmentioning

confidence: 99%

Section: Dl-based Methodsmentioning

confidence: 99%

See 1 more Smart Citation

LS-Net: fast single-shot line-segment detector

Nguyen

Jenssen

Roverso³

2020

Machine Vision and Applications

View full text Add to dashboard Cite

show abstract

“…Whenever the data is linearly separated, all neurons can have a linear activation function that maps linearly from input to output. For data that cannot be separated linearly, the algorithm will use a nonlinear activation function, such as a logistic or sigmoid function [32]. The output of this network is the final predicted value.…”

Section: Multilayer Perceptron (Mlp)mentioning

confidence: 99%

Deep Feature Vectors Concatenation for Eye Disease Detection Using Fundus Image

et al. 2021

View full text Add to dashboard Cite

Fundus image is an image that captures the back of the eye (retina), which plays an important role in the detection of a disease, including diabetic retinopathy (DR). It is the most common complication in diabetics that remains an important cause of visual impairment, especially in the young and economically active age group. In patients with DR, early diagnosis can effectively help prevent the risk of vision loss. DR screening was performed by an ophthalmologist by analysing the lesions on the fundus image. However, the increasing prevalence of DR is not proportional to the availability of ophthalmologists who can read fundus images. It can lead to delayed prevention and management of DR. Therefore, there is a need for an automated diagnostic system as it can help ophthalmologists increase the efficiency of the diagnostic process. This paper provides a deep learning approach with the concatenate model for fundus image classification with three classes: no DR, non-proliferative diabetic retinopathy (NPDR), and proliferative diabetic retinopathy (PDR). The model architecture used is DenseNet121 and Inception-ResNetV2. The feature extraction results from the two models are combined and classified using the multilayer perceptron (MLP) method. The method that we propose gives an improvement compared to a single model with the results of accuracy, and average precision and recall of 91% and 90% for the F1-score, respectively. This experiment demonstrates that our proposed deep-learning approach is effective for the automatic DR classification using fundus photo data.

show abstract

“…Various kinds of tower with different backgrounds are included. • Conductor dataset in reference [16]: This dataset contains totally 8400 images collected from visible and infrared cameras in equal quantity. To achieve multi scale recognition, images with close and far scene are included.…”

Section: Datasets For Publicmentioning

confidence: 99%

Review of data analysis in vision inspection of power lines with an in-depth discussion of deep learning technology

Liu,

Miao,

Jiang

et al. 2020

Preprint

View full text Add to dashboard Cite

The widespread popularity of unmanned aerial vehicles enables an immense amount of power lines inspection data to be collected. How to employ massive inspection data especially the visible images to maintain the reliability, safety, and sustainability of power transmission is a pressing issue. To date, substantial works have been conducted on the analysis of power lines inspection data. With the aim of providing a comprehensive overview for researchers who are interested in developing a deep-learning-based analysis system for power lines inspection data, this paper conducts a thorough review of the current literature and identifies the challenges for future research. Following the typical procedure of inspection data analysis, we categorize current works in this area into component detection and fault diagnosis. For each aspect, the techniques and methodologies adopted in the literature are summarized. Some valuable information is also included such as data description and method performance. Further, an in-depth discussion of existing deep-learning-related analysis methods in power lines inspection is proposed. Finally, we conclude the paper with several research trends for the future of this area, such as data quality problems, small object detection, embedded application, and evaluation baseline.

show abstract

Weakly supervised learning with convolutional neural networks for power line localization

Cited by 20 publications

References 13 publications

LS-Net: fast single-shot line-segment detector

LS-Net: fast single-shot line-segment detector

Deep Feature Vectors Concatenation for Eye Disease Detection Using Fundus Image

Review of data analysis in vision inspection of power lines with an in-depth discussion of deep learning technology

Contact Info

Product

Resources

About