Background Estimation and Adaptation Model with Light-Change Removal for Heavily Down-Sampled Video Surveillance Signals

Cvetkovic, S.D.; Bakker, P.G.; Sehirris, J.; de, P. H. N.

doi:10.1109/icip.2006.312602

Cited by 11 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 3 shows the performance comparisons among various schemes in terms of the accuracy rate and frame rate. It can be seen that the LL 3 -band image generated by the MDLDWT yields a superior detection rate than that of the low resolution image generated by the Haar [1], 2×2 AF [2], LDWT 9/7-based [3], and DS [4] schemes. The MDLDWT can precisely preserve local details, such as edges and singularities.…”

Section: Resultsmentioning

confidence: 99%

“…Future work can be put to apply the proposed MDLDWT to various video surveillance applications, such as object classification and descriptive behaviors of objects. Accuracy rate Process Haar-based [1] 84.76 % 557.83 FPS 2×2 AF-based [2] 83.04 % 413.99 FPS LDWT 9/7-based [3] 87.41 % 127.83 FPS DS-based [4] 85 …”

Section: Discussionmentioning

confidence: 99%

“…The improvement is based on the former structure of the 2-D LDWT. The image quality of the low resolution image can be better preserved with the MDLDWT than that of the former low resolution methods [1][2][3][4]. In the following, the proposed MDLDWT is detailed later.…”

Section: The Proposed Methodsmentioning

confidence: 99%

“…Moreover, the LL 3 -band of the MDLDWT is employed solely to reduce the image transfer computing cost and remove noise. As documented in the experimental results, the proposed method can better retain slow motion of objects than that of the former low resolution methods [1][2][3][4], and provide effective detection performance.…”

Section: Introductionmentioning

confidence: 96%

“…The change detection using inter-frame differences is a popular method for obtaining the shape information of a video object, since it is easy to implement in real-time systems. In [4], the background estimation system discriminates the foreground objects from the background by building the background model from images. It shows that a local contrast enhancement applied prior to down-sampling (DS) improves detection sensitivity.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Improved directional lifting-based discrete wavelet transform for low resolution moving object detection

Hsia

Guo

2012

2012 19th IEEE International Conference on Image Processing

View full text Add to dashboard Cite

In this study, the current state-of-the-art in moving objects segmentation for intelligent video surveillance has been surveyed. An efficient Modified Directional Lifting-based 9/7 Discrete Wavelet Transform (MDLDWT) structure is proposed to further reduce the computational cost and preserve the fine shape information in low resolution image. Although perfect moving object detection in a practical environment is a challenging task due to the vague object shape issues in the low resolution configuration, the experimental results document that the proposed low complexity MDLDWT scheme can provide more precise detection rate for multiple moving objects, and the fine shape information can be effectively preserved for the real-time video surveillance applications in both indoor and outdoor environments.Index terms-fine shape information, modified directional lifting-based discrete wavelet transform, moving object detection, discrete wavelet transform.

show abstract

Section: Resultsmentioning

confidence: 99%