Video transmission device using variable frame method
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- SOLDIA CO LTD
- Filing Date
- 2025-12-23
- Publication Date
- 2026-07-30
Smart Images

Figure KR2025022581_30072026_PF_FP_ABST
Abstract
Claims
1. A receiver that continuously receives a GOP (Group of Pictures) from a camera unit; A first conversion unit that converts an 'I' frame within each GOP received from the receiver into a first 'I' frame by performing the same conversion, converts the first 'P' frame immediately following the 'I' frame into a second 'I' frame by including image difference information of the 'P' frame itself and image reference information of the first 'I' frame, and converts each of the remaining 'P' frames after the first 'P' frame into a third 'I' frame by including image difference information of the 'P' frame itself, image difference information of at least one preceding 'P' frame, and image reference information of the first 'I' frame; A variable frame setting unit that compares each of the first, second, and third 'I' frames converted by the first conversion unit with blocks located at the same position as the immediately preceding 'I' frame between each of the first, second, and third 'I' frames, and determines a block as a valid block if the average level difference value for brightness with applied absolute values is greater than or equal to a preset first threshold value, and sets an 'I' frame as a variable target 'I' frame to be converted into a 'P' frame if the ratio of the valid block to the total block of each 'I' frame is less than or equal to a preset second threshold value; When a variable target 'I' frame is set by the variable frame setting unit, each 'I' frame is converted into a 'P' frame having image difference information compared with the immediately preceding 'I' frame, wherein the first 'I' frame among each 'I' frame is converted in the same way to a 1-1 'I' frame, and for the 'I' frame corresponding to the variable target 'I' frame, converting it into a 'P' frame having image difference information with the immediately preceding 'I' frame, thereby reconstructing the GOP by converting up to an 'I' frame greater than or equal to the second threshold value; and A video transmission device using a variable frame method, characterized by comprising a transmission unit that repeatedly transmits the GOP reconstructed by the second conversion unit to the server unit.
2. In claim 1, the second converter A video transmission device using a variable frame method, characterized by converting at least one variable target 'I' frame into a 'P' frame based on a GOP range set by a user, in a video frame received without undergoing the process of converting all frames by the first conversion unit into 'I' frames.
3. In claim 2, the second converter A video transmission device using a variable frame method, characterized in that a GOP range is set based on at least one of the specifications of the transmission unit, network traffic between the transmission unit and the server unit, and the specifications of a video receiving device that receives video from the server unit.
4. In claim 1, the second converter A video transmission device using a variable frame method, characterized by including the above-mentioned 1-1 'I' frame and reconstructing a GOP by a user so that the number of variable target 'I' frames does not exceed a preset number of frames, wherein within each reconstructed GOP, one 'I' frame having image reference information and 'P' frames having image difference information compared with the immediately preceding 'I' frame are included.
5. In claim 1, the receiving unit, the first converter, the variable frame setting unit, the second converter, and the transmitting unit A video transmission device using a variable frame method characterized by being performed in a region of interest set by the user for all frames.
6. In claim 5, the second converter A video transmission device using a variable frame method characterized by applying weights to the average level of image brightness for each specific block arbitrarily selected by the user among a plurality of blocks included in the above-mentioned region of interest.