• Khushali R. Raval
• Mahesh M. Goyani

As viewers interestingly consume sports videos significance of video summarization is increasing nowadays. In sports video summarization brief form of a sports game is constructed and the interesting segments from lengthy sports videos are highlighted. This research is undertaken to detect and classify exciting events of cricket matches like four runs, six runs and wicket fall. First, cricket video is divided into shots, and keyframes from each shot are detected. To reduce processing time all other processes are applied to keyframes. In the following stage reply segments are identified and removed from further consideration. Keyframes are classified into different views. Finally, scoreboard text is detected to identify run and wicket differences. If the run difference is equal to 4 or 6 and the wicket difference is equal to 1, then frames between such keyframes are used to build the match's highlight. The wicket fall segment of the video is further classified into a different type using text recognition based approach. The proposed technique is modest and efficient. Experiments and results reflect the effectiveness of the proposed method. The wicket classification technique using text detection and recognition is a novel contribution that achieves 100% precision and 93% recall. Proposed method found novel and generalized to classify all different types of wicket fall using a single approach.

Sports video summarization, Sports event detection, Sports video analysis, Wicket fall classification, Cricket highlight generation.

"An Event Detection based Video Summarization for Cricket Video", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 10, page no.g256-g263, October-2023, Available :http://www.jetir.org/papers/JETIR2310529.pdf

"An Event Detection based Video Summarization for Cricket Video", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 10, page no. ppg256-g263, October-2023, Available at : http://www.jetir.org/papers/JETIR2310529.pdf