Parallel Implementation of the Integral Histogram

Bellens, Pieter; Palaniappan, Kannappan; Badía, Rosa M.; Seetharaman, Guna; Labarta, Jesús

doi:10.1007/978-3-642-23687-7_53

Cited by 20 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These strips are compute in a pipeline way. For an mn  image, if the width of each strip is w (Usually m could be exactly divided by w ), there would be / mw strips, each strip requires 2 log w steps to compute the integral image [10]. In experiments, we find that when =32 w , the architecture achieves the highest power efficiency and area efficiency.…”

Section: Advances In Intelligent Systems Research Volume 130mentioning

confidence: 97%

Implementation of Face Detection System Based on ZYNQ FPGA

Feng¹,

Zheng²,

Xiao³

2016

Proceedings of the 2016 6th International Conference on Mechatronics, Computer and Education Informationization (MCEI 2016)

View full text Add to dashboard Cite

Abstract. Adaboost algorithm based on Haar-like feature and integral image is made up of cascaded classifier, and the computation is very large. This paper uses the hardware and software co-design for implementing the face detection system on Zynq-7000.And realizes the functions of video acquisition, processing, real-time detection and display, and designs the hardware acceleration module of the integral image. At the same time, it improves the detection speed of the system by using dual-core operation, and the detection speed of the system is obviously improved compared with the pure software solution. This design is verified on ZedBoard. It can work with high resolution stream for real-time face detection.

show abstract

Section: Advances In Intelligent Systems Research Volume 130mentioning

confidence: 97%

Implementation of Face Detection System Based on ZYNQ FPGA

Feng¹,

Zheng²,

Xiao³

2016

Proceedings of the 2016 6th International Conference on Mechatronics, Computer and Education Informationization (MCEI 2016)

View full text Add to dashboard Cite

show abstract

“…The integral histogram is a method to compute a cumulative histogram for each pixel of an image [7]. The OmpSs implementation performs a blocked cross-weave scan for each block of the image.…”

Section: Integral Histogrammentioning

confidence: 99%

Criticality-Aware Dynamic Task Scheduling for Heterogeneous Architectures

Chronaki

Rico

Badía

et al. 2015

Proceedings of the 29th ACM on International Conference on Supercomputing

Self Cite

View full text Add to dashboard Cite

Current and future parallel programming models need to be portable and efficient when moving to heterogeneous multicore systems. OmpSs is a task-based programming model with dependency tracking and dynamic scheduling. This paper describes the OmpSs approach on scheduling dependent tasks onto the asymmetric cores of a heterogeneous system. The proposed scheduling policy improves performance by prioritizing the newly-created tasks at runtime, detecting the longest path of the dynamic task dependency graph, and assigning critical tasks to fast cores. While previous works use profiling information and are static, this dynamic scheduling approach uses information that is discoverable at runtime which makes it implementable and functional without the need of an oracle or profiling. The evaluation results show that our proposal outperforms a dynamic implementation of Heterogeneous Earliest Finish Time by up to 1.15×, and the default breadth-first OmpSs scheduler by up to 1.3× in an 8-core heterogeneous platform and up to 2.7× in a simulated 128-core chip.

show abstract

“…Two scan orders, cross-weave and wavefront, are explored in [Bellens 2011]. The cross-weave scan processes the image in the first dimension (horizontal scan) followed by a scan in the second dimension (vertical scan).…”

Section: Integral Histogram Computationmentioning

confidence: 99%

Design Tools for Accelerating Development and Usage of Multi-Core Computing Platforms

Bhattacharyya¹

2014

View full text Add to dashboard Cite

Parallel Implementation of the Integral Histogram

Cited by 20 publications

References 15 publications

Implementation of Face Detection System Based on ZYNQ FPGA

Implementation of Face Detection System Based on ZYNQ FPGA

Criticality-Aware Dynamic Task Scheduling for Heterogeneous Architectures

Design Tools for Accelerating Development and Usage of Multi-Core Computing Platforms

Contact Info

Product

Resources

About