Block-run-based connected component labelling algorithm for GPGPU using shared memory

Chen, P.; Zhao, Hualong; Chen, Tao; Sang, Hongshi

doi:10.1049/el.2011.2941

Cited by 15 publications

(4 citation statements)

References 5 publications

(5 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Chen et al [35] proposed a two-scan approach, extended from a stripe-based CCL method [36], to process stripe extraction and stripe union, respectively.…”

Section: Line-based CCL Algorithmmentioning

confidence: 99%

Efficient Parallel Connected Component Labeling With a Coarse-to-Fine Strategy

et al. 2018

View full text Add to dashboard Cite

This paper proposes a new parallel approach to solve connected components on a 2D binary image implemented with CUDA. We employ the following strategies to accelerate neighborhood exploration after dividing an input image into independent blocks. In the local labeling stage, a coarse-labeling algorithm, including row-column connection and label-equivalence list unification, is applied first to sort out the mess of an initialized local label map; a refinement algorithm is then introduced to merge separated sub-regions from a single component. In the block merge stage, we scan the pixels located on the boundary of each block instead of solving the connectivity of all the pixels. With the proposed method, the length of label-equivalence lists is compressed, and the number of memory accesses is reduced. Thus, the efficiency of connected components labeling is improved. Experimental results show that our method outperforms the other approaches between 29% and 80% on average.

show abstract

“…Chen et al [35] proposed a two-scan approach, extended from a stripe-based CCL method [36], to process stripe extraction and stripe union, respectively.…”

Section: Line-based CCL Algorithmmentioning

confidence: 99%

Efficient Parallel Connected Component Labeling With a Coarse-to-Fine Strategy

et al. 2018

View full text Add to dashboard Cite

show abstract

“…After applying the threshold to the image, the labeling process is carried out. A state-of-the-art labeling algorithm proposed by Chen [8] has been used which scans the image to find unconnected areas that are assigned a different label. Figure 2a, b shows both a thresholded image and a labeled image, respectively.…”

Section: Case 1: Thresholding and Labeling Algorithmmentioning

confidence: 99%

Design of an accurate and high-speed binocular pupil tracking system based on GPGPUs

et al. 2017

View full text Add to dashboard Cite

An efficient and robust pupil tracking system is an important tool in visual optics and ophthalmology. It is also central to techniques for gaze tracking, of use in psychological and medical research, marketing, human-computer interaction, virtual reality and other areas. A typical setup for pupil tracking includes a camera linked to infrared LED illumination. In this work, we evaluate and parallelize several pupil tracking algorithms with the aim of accurately estimating the pupil position and size in both eyes simultaneously, to be applied in a high-speed binocular pupil tracking system. To achieve high processing speed, the original non-parallel algorithms have been parallelized by using CUDA and OpenMP. Modern graphics processors are designed to process images at high temporal frequencies and spatial resolution, and CUDA enables them to be used for general-purpose computing. Our implementation allows for efficient binocular pupil tracking at high speeds for high-resolution images (up to 988 fps with images of 1280 × 1024 pixels) using a state-of-the-art GPU. Keywords High-speed eye pupil tracking • GPU • GPGPU • Image processing • Real-time systems • Gaze tracking

show abstract

“…The number of inside-outside tests [39] or point-in-polygon checks is reduced by extracting connected components [40] from the dilated occupancy grid and using a single reference point per blob. The dilation of the grid with a circle with diameter of the vehicle width is performed to approximate the configuration space obstacles [30].…”

Section: A Environment-based Clustering With Point-in-polygon Testsmentioning

confidence: 99%

Grid-Based Multi-Road-Course Estimation Using Motion Planning

Tanzmeister

Wollherr²,

Buss³

2016

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

Knowing the course of the road together with the corresponding road boundaries is an essential component of many advanced driver assistance systems and of autonomous vehicles. This work presents an indirect, grid-based approach for road course estimation. Due to the grid representation, it is independent of specific features or particular sensors and is able to handle continuous as well as sparse road boundaries of arbitrary shape. Furthermore, the number of road courses in the scene is determined to detect road junctions and forks in the road, and the boundaries of each road course are estimated individually. The approach is based on local path planning and path clustering to find the principal moving directions through the environment. They separate the boundaries and are used for their extraction. The set of local paths and principal moving directions is reduced with approximate knowledge of the road velocity paired with system constraints, and validation and tracking assure the required robustness. Experimental results from autonomous navigation of a vehicle through an unmapped road construction site as well as quantitative evaluations demonstrate the performance of the method.

show abstract

Block-run-based connected component labelling algorithm for GPGPU using shared memory

Cited by 15 publications

References 5 publications

Efficient Parallel Connected Component Labeling With a Coarse-to-Fine Strategy

Efficient Parallel Connected Component Labeling With a Coarse-to-Fine Strategy

Design of an accurate and high-speed binocular pupil tracking system based on GPGPUs

Grid-Based Multi-Road-Course Estimation Using Motion Planning

Contact Info

Product

Resources

About