Cloud video content information extraction system based on deep learning

• Main Authors: CAI WENZHE, LIN JINYI, LI HAN, LI XIANG
• Format: Patent
• Language: Chinese; English
• Published: 14.07.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; PHYSICS; PICTORIAL COMMUNICATION, e.g. TELEVISION

The invention discloses a cloud video content information extraction system based on deep learning, which belongs to the technical field of video information extraction and comprises a video data collection module and a video image processing module. According to the invention, the video image processing module is arranged in the system and can effectively process the obtained video, so that the quality of the image is effectively improved, the accuracy of subsequent video information extraction is improved, and meanwhile, the deep learning module is arranged in the system, so that when a new round of video information extraction is carried out subsequently, the accuracy of the video information extraction is improved. The video information extraction efficiency and precision can be greatly improved, the system application effect can be improved, a video information table building module is further arranged in the system, extracted video information can be classified, class tables of different video informati