Key Frame Extraction of Surveillance Video Based on Frequency Domain Analysis

Zhang, Yunzuo; Zhang, Shasha; Zhang, Jiayu; Guo, Kaina; Cai, Zhaowei

doi:10.32604/iasc.2021.017200

Cited by 8 publications

(13 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above three comparisons indicate that the performances of the proposed method are better than the methods in [9,62,63] in a comparable or less computation time. Therefore, the effectiveness of this algorithm has been confirmed.…”

Section: Experiments and Analysismentioning

confidence: 80%

“…We can observe that the computation time of the proposed method decreases as the motion segment ratio decreases. Specifically, when the motion segment ratio is 51%, the computation time of the proposed method is significantly better than the methods in [62,63]. Meanwhile, at this ratio, the computation time of our method takes a comparable computation time with the method in [9].…”

Section: Experiments and Analysismentioning

confidence: 84%

“…From Figure 2, we can observe that the proposed method has a notable advantage in SRD. For the cases where the key frame ratio is lower than 3%, it can be seen that the comparison methods except the method in [62] have almost the same SRD. However, with the increasing key frame rate, the proposed method achieves a significant advantage of up to 1.2 db compared to the method in [9].…”

Section: Experiments and Analysismentioning

confidence: 93%

“…To verify the effectiveness of the algorithm, shot reconstruction degree [8] (SRD) is used, which is a criterion based on how well key frames capture the changes of video contents. At the same time, we compare the method proposed in this paper with the methods in literature [9, 62, 63]. The SRDs of the four compared methods on video (A) are very similar.…”

Section: Experiments and Analysismentioning

confidence: 95%

“…It is reasonably efficient and accurate. However, the proposed method is less applicable in the cases where video spatiotemporal slices are all occupied by FIGURE 2 The average SEDs of the methods in [9,62,63] and the proposed method crowded object movements all the time. Fortunately, surveillance videos are of sparse object movements in most situations [11].…”

Section: Proposed Methodsmentioning

confidence: 99%

See 4 more Smart Citations

Video key frame extraction based on scale and direction analysis

Dong¹,

Zhang

et al. 2022

The Journal of Engineering

Self Cite

View full text Add to dashboard Cite

Currently, efficient and accurate key frame extraction for massive videos remains a challenge, especially in surveillance applications. To overcome the inaccuracy of traditional key frame extraction, this paper proposes a method for key frame extraction via analyzing scale and direction of motion trajectory on spatiotemporal slice (MTSS). The proposed method employs both local and global motion state changes of object as the metric to extract key frames and meanwhile gives a strategy which integrates coarse extraction with partial fine re‐extraction of spatiotemporal slices to capture these state changes. Firstly, coarse extraction of spatiotemporal slices is employed to detect motion segments. Then partial fine re‐extraction is performed on the motion segments to obtain MTSS. Finally, the frames at the inflexions of scale and direction of MTSS are extracted as key frames. The experimental results have demonstrated that the proposed method outperforms existing state‐of‐the‐art methods in terms of accuracy for various videos while requiring a comparable or even less computation time.

show abstract

Section: Experiments and Analysismentioning

confidence: 80%

Section: Experiments and Analysismentioning

confidence: 84%