Sobel Edge Detection Based on Weighted Nuclear Norm Minimization Image Denoising

Tian, Run Lan; Sun, Guiling; Liu, Xiaochao; Zheng, Bowen

doi:10.3390/electronics10060655

Cited by 47 publications

(25 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, the traditional image filtering operator is equivalent to performing a difference operation on the pixel values in the image domain when performing operations with the image domain due to the fixed value in the operator. Therefore, traditional image filtering operators such as Sobel [17,19] can extract gradient features well, such as image edges in traditional machine vision. Therefore, due to the limitations of traditional convolution operator, there is a shortage of gradient information extraction.…”

Section: Traditional Convolutionmentioning

confidence: 99%

“…In computer vision, gradient measurement methods based on directional derivative calculation are used for image edge detection [16] to extract image features, such as Sobel operator, Robert operator, Laplacian operator, etc. [17][18][19]. In order to improve the ability of convolutional layer in gradient information extraction and further the accuracy of face detection in face anti-spoofing, this paper proposes directional difference convolution (DDC) according to the correlation across image pixels.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Directional Difference Convolution and Its Application on Face Anti-Spoofing

Yang

Zhao

et al. 2022

Mathematics

View full text Add to dashboard Cite

In practical application, facial image recognition is vulnerable to be attacked by photos, videos, etc., while some currently used artificial feature extractors in machine learning, such as activity detection, texture descriptors, and distortion detection, are insufficient due to their weak detection ability in feature extraction from unknown attack. In order to deal with the aforementioned deficiency and improve the network security, this paper proposes directional difference convolution for the deep learning in gradient image information extraction, which analyzes pixel correlation within the convolution domain and calculates pixel gradients through difference calculation. Its combination with traditional convolution can be optimized by a parameter θ. Its stronger ability in gradient extraction improves the learning and predicting ability of the network, whose performance testing on CASIA-MFSD, Replay-Attack, and MSU-MFSD for face anti-spoofing task shows that our method outperforms the current related methods.

show abstract

Section: Traditional Convolutionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Directional Difference Convolution and Its Application on Face Anti-Spoofing

Yang

Zhao

et al. 2022

Mathematics

View full text Add to dashboard Cite

show abstract

“…In order to further improve the accuracy of the segmentation mask, a method of adding edge loss [ 22 ] to the mask branch is proposed to make the edge of the segmentation result more accurate. First, the labeled image is converted into a binary segmentation map of the crop, which is the target mask, and then the prediction mask and the target mask output by the mask branch are used as input, and they are convolved with the Sobel operator [ 23 ]. The Sobel operator is a two-dimensional operator such as Equations (3) and (4).…”

Section: Improved Mask Rcnn Algorithm Based On Crop Image Extractionmentioning

confidence: 99%

A Crop Image Segmentation and Extraction Algorithm Based on Mask RCNN

Wang

Sun

Zheng

et al. 2021

Entropy

Self Cite

View full text Add to dashboard Cite

The wide variety of crops in the image of agricultural products and the confusion with the surrounding environment information makes it difficult for traditional methods to extract crops accurately and efficiently. In this paper, an automatic extraction algorithm is proposed for crop images based on Mask RCNN. First, the Fruits 360 Dataset label is set with Labelme. Then, the Fruits 360 Dataset is preprocessed. Next, the data are divided into a training set and a test set. Additionally, an improved Mask RCNN network model structure is established using the PyTorch 1.8.1 deep learning framework, and path aggregation and features are added to the network design enhanced functions, optimized region extraction network, and feature pyramid network. The spatial information of the feature map is saved by the bilinear interpolation method in ROIAlign. Finally, the edge accuracy of the segmentation mask is further improved by adding a micro-fully connected layer to the mask branch of the ROI output, employing the Sobel operator to predict the target edge, and adding the edge loss to the loss function. Compared with FCN and Mask RCNN and other image extraction algorithms, the experimental results demonstrate that the improved Mask RCNN algorithm proposed in this paper is better in the precision, Recall, Average precision, Mean Average Precision, and F1 scores of crop image extraction results.

show abstract

“…When sailing in east-west direction, the vertical edge-detection operator is used to divide the suitable area; the horizontal edgedetection operator is used when sailing in north-south direction; the 45 • edge-detection operator is used for northwest direction, and the 135 • edge-detection operator is used for northeast direction. The Sobel operators are as shown in Figure 1 [15,16]. The calculation of each pixel in the feature map is shown in Formula (1).…”

Section: Image Convolutionmentioning

confidence: 99%

Optimizing Matching Area for Underwater Gravity-Aided Inertial Navigation Based on the Convolution Slop Parameter-Support Vector Machine Combined Method

Wang

Zheng

2021

Remote Sensing

View full text Add to dashboard Cite

This paper focuses on the selection of matching areas in the gravity-aided inertial navigation system. Firstly, the Sobel operator was used in convolution of the gravity anomaly map to obtain the feature map. The convolution slope parameters were constructed by combining the feature map and the gravity anomaly map. The characteristic parameters, such as the difference between convolution rows and columns, convolution variance of the feature map, the pooling difference, and range of the gravity anomaly map, were combined. Based on the support vector machine algorithm, the convolution slope parameter–support vector machine combined method is proposed. Second, we selected the appropriate training sample set and set parameters to verify. The results show that compared with the pre-calibration results, the classification accuracy of the test set is more than 92%, which proves that the convolution slope parameter–support vector machine combined method can effectively distinguish between the suitable and the unsuitable area. Thirdly, we applied this method to another region. The navigation experiment was performed in the split-matching area. The average positioning error was better than 100 m, and the correct rate was more than 90%. The results show that sailing in the selected area can accurately match the trajectory and reduce the positioning error.

show abstract

Sobel Edge Detection Based on Weighted Nuclear Norm Minimization Image Denoising

Cited by 47 publications

References 19 publications

Directional Difference Convolution and Its Application on Face Anti-Spoofing

Directional Difference Convolution and Its Application on Face Anti-Spoofing

A Crop Image Segmentation and Extraction Algorithm Based on Mask RCNN

Optimizing Matching Area for Underwater Gravity-Aided Inertial Navigation Based on the Convolution Slop Parameter-Support Vector Machine Combined Method

Contact Info

Product

Resources

About