Fast Background Subtraction Based on GPU

2011 ◽  
Vol 341-342 ◽  
pp. 737-742
Author(s):  
Jian Ping Han ◽  
Xiao Yang Li ◽  
Da Xing Zhang ◽  
Bo Ting Geng
Keyword(s):  
Background Subtraction ◽  
Moving Objects ◽  
Experimental Results ◽  
Surveillance Videos ◽  
Effectiveness And Efficiency ◽  
Asynchronous Execution

In this paper, a fast background subtraction algorithm using codebook model is presented to extract moving objects from surveillance videos. The time for stopped objects being absorbed into the background can be controlled to deal with different applications and have nothing to do with the complexity of the scene. We implement the algorithm on GPU using CUDA, and optimize the implementation using pinned memory and asynchronous execution techniques. Experimental results are provided to demonstrate the accuracy, effectiveness, and efficiency of the proposed algorithm.

Foggy Port Surveillance Video Clearness Algorithm Based on Wavelet Packet Decomposition

2014 ◽  
Vol 644-650 ◽  
pp. 4616-4619
Author(s):  
Zhi Yuan Xu ◽  
Yong Kai Wang ◽  
Xiao Hong Su ◽  
Yi Liu
Keyword(s):  
Moving Objects ◽  
Wavelet Packet ◽  
Experimental Results ◽  
Wavelet Packet Decomposition ◽  
Surveillance Video ◽  
Surveillance Videos ◽  
Background Image

Port surveillance videos are degraded seriously in foggy conditions. This paper presented a clearness algorithm based on wavelet packet decomposition. Firstly, we extracted the background image from degraded videos and established the updated model; Secondly, we detected the moving objects as foreground images; Thirdly, we defogged these images based on wavelet packet decomposition; Finally, we fused the background and foreground images together. The experimental results show that our method is more effective.

Intelligent Monitoring of Key Sections in Public Transportation

2014 ◽  
Vol 722 ◽  
pp. 353-358
Author(s):  
Yong Wu Wu ◽  
You Fu Wu ◽  
Zu Feng Fu ◽  
Shu Qu Qian
Keyword(s):  
Behavior Analysis ◽  
Hough Transform ◽  
Public Transportation ◽  
Background Subtraction ◽  
Moving Objects ◽  
Experimental Results ◽  
Advanced Stage ◽  
Polynomial Fitting ◽  
Intelligent Monitoring ◽  
Hu Moment

Behavior analysis is the advanced stage in intelligence surveillance. In this paper, we first parameterize the scene knowledge using Hough Transform and Polynomial fitting to boundary of road. The algorithm of Self-Adaptive Background Subtraction was cited in order to segment the moving objects; the features of improved Hu moment were used for classification; and the cordon was cited to realize the behavior analysis of moving objects. The experimental results show that our algorithm is effective.

A Real Implementation of Dangerous Objects Capturing in Surveillance Web

2013 ◽  
Vol 278-280 ◽  
pp. 1032-1035
Author(s):  
Yun Cheng ◽  
Hai Tao Lang ◽  
Peng Yao ◽  
Rui Guo ◽  
Jian Ying Hu ◽  
...  
Keyword(s):  
Background Subtraction ◽  
Moving Objects ◽  
Texture Feature ◽  
Multiple Scale ◽  
Experimental Results ◽  
Phase Method ◽  
Two Phase ◽  
Surveillance Camera ◽  
Object Learning ◽  
Classification Framework

The main focus of our research is capturing dangerous objects when they appear under the surveillance camera again, while have performed a dangerous activities in other places. Our solution is a two-phase method, including object learning and capturing under the classification framework. The samples of objects and non-objects are collected to train a classifier with libSVM in object learning phase. In object capturing phase, all moving objects are detected by background subtraction, then are classified into dangerous or non-dangerous. To obtain a robust objects representation to illumination, scale, rotation etc. we fuse HSV space based color feature and multiple scale texture feature. The experimental results with real surveillance data validated the proposed method.

scholarly journals Background Modelling using a Q-Tree Based Foreground Segmentation

2020 ◽  
Vol 21 (1) ◽  
pp. 17-31
Author(s):  
S Shahidha Banu ◽  
N Maheswari
Keyword(s):  
Background Subtraction ◽  
Moving Objects ◽  
False Negative ◽  
Region Of Interest ◽  
Intensity Variation ◽  
Experimental Results ◽  
Foreground Segmentation ◽  
Foreground Detection ◽  
Foreground Object ◽  
Background Modelling

Background modelling is an empirical part in the procedure of foreground mining of idle and moving objects. The foreground object detection has become a challenging phenomenon due to intermittent objects, intensity variation, image artefact and dynamic background in the video analysis and video surveillance applications. In the video surveillances application, a large amount of data is getting processed by everyday basis. Thus it needs an efficient background modelling technique which could process those larger sets of data which promotes effective foreground detection. In this paper, we presented a renewed background modelling method for foreground segmentation. The main objective of the work is to perform the foreground extraction only inthe intended region of interest using proposed Q-Tree algorithm. At most all the present techniques consider their updates to the pixels of the entire frame which may result in inefficient foreground detection with a quick update to slow moving objects. The proposed method contract these defect by extracting the foreground object by controlling the region of interest (the region only where the background subtraction is to be performed) and thereby reducing the false positive and false negative. The extensive experimental results and the evaluation parameters of the proposed approach with the state of art method were compared against the most recent background subtraction approaches. Moreover, we use challenge change detection dataset and the efficiency of our method is analyzed in different environmental conditions (indoor, outdoor) from the CDnet2014 dataset and additional real time videos. The experimental results were satisfactorily verified the strengths and weakness of proposed method against the existing state-of-the-art background modelling methods.

scholarly journals Detection of Moving Objects with Fuzzy Color Coherence Vector

2014 ◽  
Vol 2014 ◽  
pp. 1-8 ◽  
Author(s):  
Yu-Long Qiao ◽  
Kai-Long Yuan ◽  
Chun-Yan Song ◽  
Xue-Zhi Xiang
Keyword(s):  
Background Subtraction ◽  
Moving Objects ◽  
Spatial Relationship ◽  
Experimental Results ◽  
Fuzzy Membership ◽  
Subtraction Method ◽  
Color Distribution ◽  
Popular Method ◽  
Statistical Feature ◽  
Advanced Applications

Background subtraction is a popular method for detecting foreground that is widely adopted as the fundamental processing for advanced applications such as tracking and surveillance. Color coherence vector (CCV) includes both the color distribution information (histogram) and the local spatial relationship information of colors. So it overcomes the weakness of the conventional color histogram for the representation of an object. In this paper, we introduce a fuzzy color coherence vector (FCCV) based background subtraction method. After applying the fuzzyc-means clustering to color coherence subvectors and color incoherence subvectors, we develop a region-based fuzzy statistical feature for each pixel based on the fuzzy membership matrices. The features are extracted from consecutive frames to build the background model and detect the moving objects. The experimental results demonstrate the effectiveness of the proposed approach.

A Background Subtraction Method for Defect Detection of Printed Image

2013 ◽  
Vol 462-463 ◽  
pp. 421-427
Author(s):  
Jian Hua Ding ◽  
Yao Lu ◽  
Wei Huang ◽  
Ming Qin
Keyword(s):  
Defect Detection ◽  
Background Subtraction ◽  
Moving Objects ◽  
Texture Feature ◽  
Background Modeling ◽  
Experimental Results ◽  
Subtraction Method ◽  
Moving Objects Detection ◽  
Hsv Histogram ◽  
Objects Detection

Background subtraction is often used to detect the moving objects from static cameras. The difficult of defect detecting of printing matter is how to detect the unknown flaws in complicate background effectively. Inspired by the background modeling approach for moving objects detection, a background modeling method in defect detection of printed image is suggested in this paper. Those pixels without defects are regarded as background, while the flaw pixels are defined as foreground. Firstly, we select LBP histogram as texture feature and HSV histogram as color feature to model the background respectively. Then, lots of printed images in which there are no defects are used to update these two models. Finally, we utilize the models to detect defects of printing images. Experimental results show that this background model works well in our defect detection.

scholarly journals Object-Based Approach for Adaptive Source Coding of Surveillance Video

2019 ◽  
Vol 9 (10) ◽  
pp. 2003 ◽  
Author(s):  
Tung-Ming Pan ◽  
Kuo-Chin Fan ◽  
Yuan-Kai Wang
Keyword(s):  
Linear Model ◽  
Video Transmission ◽  
Moving Objects ◽  
Source Coding ◽  
Video Quality ◽  
Surveillance Video ◽  
Surveillance Videos ◽  
Object Based ◽  
High Recognition Accuracy ◽  
Object Dynamics

Intelligent analysis of surveillance videos over networks requires high recognition accuracy by analyzing good-quality videos that however introduce significant bandwidth requirement. Degraded video quality because of high object dynamics under wireless video transmission induces more critical issues to the success of smart video surveillance. In this paper, an object-based source coding method is proposed to preserve constant quality of video streaming over wireless networks. The inverse relationship between video quality and object dynamics (i.e., decreasing video quality due to the occurrence of large and fast-moving objects) is characterized statistically as a linear model. A regression algorithm that uses robust M-estimator statistics is proposed to construct the linear model with respect to different bitrates. The linear model is applied to predict the bitrate increment required to enhance video quality. A simulated wireless environment is set up to verify the proposed method under different wireless situations. Experiments with real surveillance videos of a variety of object dynamics are conducted to evaluate the performance of the method. Experimental results demonstrate significant improvement of streaming videos relative to both visual and quantitative aspects.

Multi Object Detection and Tracking from Video File

2014 ◽  
Vol 533 ◽  
pp. 218-225 ◽  
Author(s):  
Rapee Krerngkamjornkit ◽  
Milan Simic
Keyword(s):  
Computer Vision ◽  
Background Subtraction ◽  
Moving Objects ◽  
Gaussian Mixture Models ◽  
Gaussian Mixture ◽  
Video Tracking ◽  
Partial Occlusion ◽  
Video File ◽  
Detection And Tracking ◽  
Over Time

This paper describes computer vision algorithms for detection, identification, and tracking of moving objects in a video file. The problem of multiple object tracking can be divided into two parts; detecting moving objects in each frame and associating the detections corresponding to the same object over time. The detection of moving objects uses a background subtraction algorithm based on Gaussian mixture models. The motion of each track is estimated by a Kalman filter. The video tracking algorithm was successfully tested using the BIWI walking pedestrians datasets [. The experimental results show that system can operate in real time and successfully detect, track and identify multiple targets in the presence of partial occlusion.

Algorithm of Color Detection for Moving Video Objects Based on Mode Matching

2012 ◽  
Vol 239-240 ◽  
pp. 1000-1003
Author(s):  
Zhao Quan Cai ◽  
Hui Hu ◽  
Tao Xu ◽  
Wei Luo ◽  
Yi Cheng He
Keyword(s):  
Background Subtraction ◽  
Moving Objects ◽  
Color Matching ◽  
Human Beings ◽  
Current Image ◽  
Color Recognition ◽  
Recognition Ability ◽  
Color Identification ◽  
The Difference ◽  
Background Image

It is urgent to study how to effectively identify color of moving objects from the video in the information era. In this paper, we present the color identification methods for moving objects on fixed camera. One kind of the methods is background subtraction that recognizes the foreground objects by compare the difference of pixel luminance between the current image and the background image at the same coordinates. Another kind is based on the statistics of HSV color and color matching which makes the detection more similar to the color identification of the human beings. According to the experiment results, after the completion of the background modelling, our algorithm of background subtraction, statistics of the HSV color and the color matching have strong color recognition ability on the moving objects of video.

Dynamic world modeling for a mobile robot among moving objects

Robotica ◽  
1996 ◽  
Vol 14 (5) ◽  
pp. 553-560
Author(s):  
Yuefeng Zhang ◽  
Robert E. Webber
Keyword(s):  
Mobile Robot ◽  
Hough Transform ◽  
Moving Objects ◽  
Grid Method ◽  
Experimental Results ◽  
Sensor Data ◽  
Probabilistic Estimation ◽  
Occupancy Grid ◽  
World Modeling ◽  
Grid Based

SUMMARYA grid-based method for detecting moving objects is presented. This method involves the extension and combination of two methods: (1) the Hough Transform and (2) the Occupancy Grid method. The Occupancy Grid method forms the basis for a probabilistic estimation of the location and velocity of objects in the scene from the sensor data. The Hough Transform enables the new method to handle non-integer velocity values. A model for simulating a sonar ring is also presented. Experimental results show that this method can handle objects moving at non-integer velocities.

Sign in / Sign up

Export Citation Format

Share Document