GUD-Canny: a real-time GPU-based unsupervised and distributed Canny edge detector

Fuentes-Alventosa, Antonio; Gómez-Luna, Juan; Medina-Carnicer, R.

doi:10.1007/s11554-022-01208-0

Cited by 8 publications

(7 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the characteristics of FPGA parallel operation, the convolution algorithm is replaced by the template, and the system structure is improved and optimized by taking advantage of the pipeline, thus, the real-time video edge detection system architecture is obtained in reference [1]; Based on the characteristics of FPGA programmable chips, reference [2] encapsulated the edge detection algorithm based on Canny into a IP core, a the real-time edge detection of FPGA is realized, and the development cycle of FPGA is shortened as well. An unsupervised distributed Canny edge detection algorithm implemented on GPU is proposed in reference [3] in order to deal with the detection bottleneck caused by fixed threshold and lag.…”

Section: Principle Overview Of Canny Edge Detection Algorithmmentioning

confidence: 99%

Research on Image Edge Detection Algorithm Based on CORDIC

Wang

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Image edge extraction is the foundation of high-level image processing. In view of the characteristics of large amount of square root algorithm and difficult hardware implementation in traditional Canny edge detection, the CORDIC algorithm is proposed in this paper, and the square root operation is completed by shifting, adding and subtracting, and multiplying. According to the experiment, under the premise of limited resources on FPGA, the image edge extraction can be completed in real time by this method, which opens up a new idea for real-time image processing.

show abstract

Section: Principle Overview Of Canny Edge Detection Algorithmmentioning

confidence: 99%

Research on Image Edge Detection Algorithm Based on CORDIC

Wang

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

“…[42]. Inspired by citations [40] and [41], this work makes improvements to the Canny edge detection algorithm. For the contour characteristics of industrial parts, the curvature information of the parts is increased by adding two tilt direction templates to make the edge detection more accurate.…”

Section: Edge Detectionmentioning

confidence: 99%

“…Fuentes‐Alventosa et al. proposed an unsupervised distributed Canny edge detector based on GPU to meet the edge detection real‐time requirements [40]. Yang proposed a method for improving Canny edge detection by geo‐information mapping [41].…”

Section: Related Workmentioning

confidence: 99%

“…Recent edge detection algorithms focus on the accurate detection of object boundaries, and in this work, we choose to improve on the Canny algorithm. Fuentes-Alventosa et al proposed an unsupervised distributed Canny edge detector based on GPU to meet the edge detection real-time requirements[40]. Yang proposed a method for improving Canny edge detection by geo-information mapping[41].…”

mentioning

confidence: 99%

See 1 more Smart Citation

A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification

An,

Sun,

Song

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

Traditional warehousing typically needs machine learning or manual tagging to classify objects. However, this method is less robust and consumes a lot of labour and material resources. Based on DenseNet, this work proposes a feature weighting convolutional network recognition model and designs a set of software and hardware for data acquisition, which is applied to the efficient classification of industrial parts in warehouse management. Firstly, this work modifies DenseNet by embedding SE‐Block, and replaces the cross‐entropy loss function with the focus loss function to optimize the model structure. Secondly, a multi‐view hardware and software acquisition system is designed to complete the functions of part image acquisition, image preprocessing, model training and part recognition. Finally, an industrial parts sorting experiment was designed. Compared with the original DenseNet model, the proposed weighted convolutional network identification model showed that the accuracy of the modified model was increased by 3.09% and the convergence rate was significantly improved. The modified model proposed in this work aims to improve the recognition accuracy of industrial parts in modern warehouse management, so as to modify the classification efficiency of warehouse parts in production.

show abstract

“…The Canny edge detection algorithm was then applied to obtain the edge contour information in the image. The effective information and the noisy information were distinguished, and the noise was removed according to the perimeter of the contour [20]. The processing effect is presented in Figure 5.…”

Section: Waste Image Sample Preprocessingmentioning

confidence: 99%

Design and Simulation of Small-Scale Waste Separation and Sorting Equipment

Zhang

2022

Processes

View full text Add to dashboard Cite

Environmental pollution is currently a major concern globally owing to increase in the world population and advances in industrialization. An automatic waste separation and sorting equipment for small-scale waste sorting systems was constructed in this paper. A two-DoF (two-degree-of-freedom) parallel sorting mechanism as well as a belt waste separation and feeding mechanism were designed, and 3D models of the mechanism were established. Finite element analysis (FEA) was conducted for the sorting mechanism, and kinematic and dynamic analysis and numerical simulation were performed for the waste and sorting mechanism. The simulation results demonstrated that the spherical waste did not roll out of the platform after leaving the slideway. The waste was stabilized at the center of the platform within a short duration with movement of the collecting mechanism of the platform. The system was optimized to reduce waste separation and sorting time. Stability of the waste motion in the equipment as well as efficiency of the system were significantly improved after optimization. The results showed that the designed equipment can complete separation and sorting of waste materials at lower output power.

show abstract

GUD-Canny: a real-time GPU-based unsupervised and distributed Canny edge detector

Cited by 8 publications

References 44 publications

Research on Image Edge Detection Algorithm Based on CORDIC

Research on Image Edge Detection Algorithm Based on CORDIC

A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification

Design and Simulation of Small-Scale Waste Separation and Sorting Equipment

Contact Info

Product

Resources

About