Using the CPU and GPU for real-time video enhancement on a mobile computer

This work aims to evaluate the improvement in the performance of tracking small maritime targets due to real-time enhancement of the video streams from high zoom cameras on pan-tilt pedestal. Due to atmospheric conditions these images can frequently have poor contrast, or exposure of the target if it is far and thus small in the camera's field of view. A 300mm focal length lens and machine vision camera were mounted on a pan-tilt unit and used to observe the False Bay near Simon's Town, South Africa. A ground truth data-set was created by performing a least squares geo-alignment of the camera system and placing a differential global position system receiver on a target boat, thus allowing the boat's position in the camera's field of view to be determined. Common tracking techniques including level-sets, Kalman filters and particle filters were implemented to run on the central processing unit of the tracking computer. Image enhancement techniques including multi-scale tone mapping, interpolated local histogram equalisation and several sharpening techniques were implemented on the graphics processing unit. This allowed the 1.3 mega-pixel 20 frames per second video stream to be processed in real-time. A quantified measurement of each tracking algorithm's robustness in the presence of sea-glint, low contrast visibility and sea clutter -such as white caps is performed on the raw recorded video data. These results are then compared to those obtained using data enhanced with the algorithms described.

show abstract

“…Two methods are outlined below. More detail can be found in the paper on their real-time implementation 16 .…”

Section: Tone Mappingmentioning

confidence: 99%

Quantitative analysis of the improvement in high zoom maritime tracking due to real-time image enhancement

et al. 2011

View full text Add to dashboard Cite

show abstract

“…Load balancing techniques between these two units can be used according to the needs, thus the CPU can respond to other requests of the user/device while the GPU carries out another general purpose processing [13].…”

Section: Gpu Acceleration On Embedded Devicesmentioning

confidence: 99%

GPU Acceleration on Embedded Devices. A Power Consumption Approach

Calandrini

Gardel

Revenga

et al. 2012

2012 IEEE 14th International Conference on High Performance Computing and Communication &Amp; 2012 IEEE 9th International Confe

View full text Add to dashboard Cite

This paper analyses the power consumption of hybrid computation on embedded architectures with an available GPU. Novel efficiency metrics are obtained using a well-known benchmark process based on the Fourier transform as computing work load. The measurement process is arranged in order to obtain specific power data for each hardware configuration, varying the data size and number of computation threads, disabling the GPU, mixing the power computation of CPU/GPU, etc. The resulting data may be of interest for new applications and cluster development (i.e. Beowulf clusters) based on low power devices, such as the Beobot project.

show abstract

“…Singhal et al developed and optimized image processing algorithms on handheld GPU [32]. Bachoo developed real-time video enhancement algorithm on a mobile GPU computer [33]. López et al developed interactive multi-frame reconstruction for mobile GPU [34].…”

mentioning

confidence: 99%

“…Mobile GPUs are successfully applied to develop the applications of pixel processing [30][31][32][33][34]. Nitin et al developed and optimized image processing algorithms on mobile GPU [30].…”

mentioning

confidence: 99%

Fast filter bank convolution for three-dimensional wavelet transform by shared memory on mobile GPU computing

Zhao

2015

J Supercomput

View full text Add to dashboard Cite

Mobile GPU applications usually constrain by the real-time requirement. However, FLOPS of mobile GPU is limited by the size and power supply of the SoC systems. Same to desktop GPUs, the mobile GPU consists of an on-chip memory hierarchy, and proper usage of memory hierarchy accelerates mobile GPU applications such as Discrete Wavelet Transform (DWT) to satisfy the real-time requirement. In this paper, by taking advantage of GPU shared memory in Tegra K1, a mobile GPU from Nvidia, we develop Bank Conflict Free Shared Memory Parallel DWT for mobile GPU applications. Computational results show that, with the display resolution of 640 × 350 (EGA), Bank Conflict Free Shared Memory Parallel DWT is significantly faster than SoC CPU-based DWT. Computational results also show that, with the display resolution of 320 × 200 (CGA), 640 × 480 (VGA), 800 × 600 (SVGA) and 1024 × 768 (XGA), Bank Conflict Free Shared Memory Parallel DWT can generally satisfy the real-time requirement.

show abstract

Using the CPU and GPU for real-time video enhancement on a mobile computer

Cited by 7 publications

References 7 publications

Quantitative analysis of the improvement in high zoom maritime tracking due to real-time image enhancement

Quantitative analysis of the improvement in high zoom maritime tracking due to real-time image enhancement

GPU Acceleration on Embedded Devices. A Power Consumption Approach

Fast filter bank convolution for three-dimensional wavelet transform by shared memory on mobile GPU computing

Contact Info

Product

Resources

About