Video key frame extraction method, system and device based on multi-view features

Abstract

The invention discloses a video key frame extraction method, system and device based on multi-view features, and the method comprises the steps: setting a sampling rate to sample an original video stream, and extracting the video stream into a plurality of frames of images; calculating a Hamming distance of hash values of every two continuous adjacent frames of images for all the extracted framesby applying an average hash method AHA, if the Hamming distance is greater than a threshold value, determining that a shot boundary is formed, and otherwise, not dividing the shot boundary; respectively extracting three characteristic values, namely an RGB characteristic value, an HSV characteristic value and an LBP characteristic value, from each frame of image extracted in the sampling step; performing single-core clustering calculation on the extracted RGB, HSV and LBP feature values in each lens according to a lens division result in the lens division step, performing summation operation after normalization processing on clustering results, and taking a frame with a minimum summation result as a key frame of the lens. The extracted key frame is more representative, the robustness of the algorithm is enhanced, and the readability of video abstract extraction is improved.

Description

technical field [0001] The present disclosure relates to the fields of video key frame extraction and automatic generation of video summaries, in particular to a video key frame extraction method, system and device based on multi-view features. Background technique [0002] The statements in this section merely mention background art related to the present disclosure and do not necessarily constitute prior art. [0003] In the process of realizing the present disclosure, the inventors found that the following technical problems existed in the prior art: [0004] Video data is a typical unstructured data. Its data model is unclear and its data structure is irregular, making it more difficult to standardize than structured data, which determines that general data management methods cannot effectively retrieve and process it. , such as video summarization based on keyframe extraction. At present, in the field of video key frame extraction, the use of clustering algorithms to ...

AI Technical Summary

Problems solved by technology

[0005] (1) It is necessary to rely on the threshold input to define the number of clustering cores. The threshold size will directly or indirectly determine the number of key frame extractions, thereby affecting the key frame extraction effect;

[0006] (2) Treating each frame after extracting feature values ​​as the same basic element without difference, resulting in disordered sequence relationship between frames, and the formed video summary cannot reflect the plot development of the original video content;

[0007] (3) Using the color histogram of the frame image as the image feature value, ignoring the feature information such as the outline, brightness, and saturation of the image, resulting in insufficient feature value extraction

[0008] The above problems affect the reliability of key frame extraction to varying degrees, and fundamentally determine that the general key frame extraction method has shortcomings such as single applicable scene and poor robustness.

Images