Image Encoding Device Completing Group of Pictures After Stop
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image encoding devices cannot encode all obtained images, particularly B pictures, as they require subsequent images for encoding, leading to incomplete encoding when a stop instruction is received.
Innovation Solution
An image encoding device with a configuration that includes acquisition units, a switch unit, memory, and an encoding unit, which intermittently acquires and stores images, allowing for encoding in a predetermined order and using the last acquired image to complete the Group of Pictures (GOP), ensuring all images are encoded, even if the process is stopped.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If B pictures are encoded using bidirectional prediction requiring both previous and forward encoded pictures, then encoding efficiency is improved, but incomplete encoding occurs when stop instruction is received
Solution Approach 1:
The patent applies preliminary action by acquiring the subsequent picture (forward reference picture) before the stop instruction is received and storing it in advance. This allows the B picture encoding to be completed even after the stop instruction, as both the previous picture (from buffer memory) and forward picture (from subsequent acquisition) are already available. The encoding unit then performs the bidirectional prediction encoding using these pre-acquired reference pictures, ensuring complete encoding without requiring further picture acquisition after stopping.
2Speed
If the encoding process is stopped immediately when a stop instruction is received, then response time is improved, but some acquired images cannot be encoded
Solution Approach 1:
The patent applies preliminary action by continuing to acquire pictures for a predetermined period after the stop instruction is received. This ensures that the subsequent picture needed for B picture encoding is obtained in advance and stored in the buffer memory or picture storage unit. The encoding process is then completed using these pre-acquired pictures, preventing loss of any acquired images while maintaining fast response by not delaying the stop signal processing.
Solution Approach 2:
The patent applies continuity of useful action by maintaining picture acquisition and storage operations for a predetermined period after the stop instruction is received. This continuous acquisition ensures that all necessary reference pictures are available for completing the encoding of B pictures that require bidirectional prediction. The useful action of picture acquisition continues just long enough to ensure complete encoding, then stops, preventing both image loss and unnecessary extended operation.
3Reliability
If all acquired images are encoded by acquiring subsequent pictures after stop instruction, then encoding completeness is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by acquiring the subsequent picture (forward reference picture) in advance, before the stop instruction is received or immediately after, and storing it in the buffer memory. This pre-acquisition eliminates the need to wait for or acquire additional pictures after stopping, allowing B picture encoding to be completed immediately using the pre-acquired reference pictures. This resolves the contradiction by ensuring encoding completeness without extending processing time, as all necessary pictures are already available when encoding begins.
Data Source
Figure 1A~1B
Figure 2
Figure 3A~3C
AI summary
An image coding device (100) comprises a first acquisition unit (1) and a second acquisition unit (2) which each acquire a moving image, a coding unit (5) which codes the acquired image in accordance with a predetermined coding order and a predetermined coding method, and an acceptance unit (6) which accepts an instruction to stop the execution of processing, and when the acceptance unit (6) accepts the instruction, the coding unit (5) codes all of acquired images using an image last acquired as an image that should have been acquired after the image last acquired.