Image retrieval using a scale-invariant feature transform bag-of-features model with salient object detection

Shiue, Yih Chearng; Lo, Sheng Hung; Tian, Yi; Lin, Cheng

doi:10.1504/ijass.2017.088903

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The feature learning-based methods acquires the position and contours of the vehicle in three steps: training the vehicle features in whole or in part, sliding over each frame with a fixed size window, and judging whether a part is the vehicle or background. Shiue et al [13] described vehicles with fixed features, and developed a detection model with invariant features. Baek et al [14] trained the detection model with adaptive boosting (AdaBoost) algorithm, and detected moving vehicles in the binary graph.…”

Section: Moving Vehicle Detection Methodsmentioning

confidence: 99%

Moving Vehicle Detection and Tracking Based on Video Sequences

Wang¹

2020

View full text Add to dashboard Cite

Vehicle detection and tracking are key aspects of intelligent transportation. However, the accuracy of traditional methods for vehicle detection and tracking is severely suppressed by the complexity of road conditions, occlusion, and illumination changes. To solve the problem, this paper initializes the background model based on multiple frames with a fixed interval, and then establishes a moving vehicle detection algorithm based on trust interval. The established algorithm can easily evaluate the background complexity based on regional information. After that, the authors pointed out that the correlation filtering, a classical vehicle tracking algorithm, cannot adapt to the scale changes of vehicles, due to the weak dependence of background information. Hence, a tracking algorithm that adapts to vehicle scale was designed based on background information. Finally, the proposed algorithms were proved feasible for detection and tracking moving vehicles in complex environments. The research provides a good reference for the application of computer vision in moving target detection.

show abstract

Section: Moving Vehicle Detection Methodsmentioning

confidence: 99%

Moving Vehicle Detection and Tracking Based on Video Sequences

Wang¹

2020

View full text Add to dashboard Cite

show abstract

Image retrieval using a scale-invariant feature transform bag-of-features model with salient object detection

Cited by 1 publication

References 0 publications

Moving Vehicle Detection and Tracking Based on Video Sequences

Moving Vehicle Detection and Tracking Based on Video Sequences

Contact Info

Product

Resources

About