Image Processing Device Circuit Cost Reduction via Time-Division Multiplexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing devices require multiple decoding and scaling circuits to process multiple bitstreams simultaneously, leading to increased cost and power consumption, and time division multiplexing limits processing time, necessitating higher operating frequencies.
Innovation Solution
An image processing device and method that includes a decoding circuit, a computing circuit, a frame composer, and an image pre-mixer, which decodes bitstreams, generates intermediate images, and mixes them with buffered images using reference data to produce a pre-mixed image, reducing the need for multiple circuits and lowering operational frequency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple decoding circuits and scaling circuits are used to process multiple bitstreams simultaneously, then the processing capability is improved, but the circuit cost and power consumption increase
Solution Approach 1:
The patent merges multiple decoding circuits and scaling circuits into a single shared circuit by implementing time-division multiplexing. The decoding circuit decodes different bitstreams at different time slots, and the scaling circuit processes decoded frames from different bitstreams sequentially, thereby reducing the total number of circuits while maintaining the ability to process multiple bitstreams simultaneously
Solution Approach 2:
The patent implements periodic time-division multiplexing where the shared decoding circuit and scaling circuit alternate between different bitstreams in periodic time slots. Each bitstream is assigned specific time slots for decoding and scaling operations, enabling the single circuit to handle multiple bitstreams efficiently without conflict
2Quantity of substance
If time division multiplexing is used to share decoding and scaling circuits, then the circuit cost is reduced, but the processing time for each bitstream is limited
Solution Approach 1:
The patent applies preliminary action by performing motion compensation and motion estimation in advance during the decoding phase. The motion compensation unit generates predicted blocks using reference frames before the scaling phase, so that when the scaling circuit processes the decoded frame, the prediction data is already prepared, reducing the actual processing time during scaling
Solution Approach 2:
The patent ensures continuity of useful action by overlapping operations between different bitstreams. While the decoding circuit is decoding a current bitstream, the scaling circuit is simultaneously scaling previously decoded frames from other bitstreams. This continuous utilization of circuits eliminates idle time and maintains high processing throughput
3Speed
If the operating frequency is increased to compensate for limited processing time in TDM, then the processing speed is improved, but the power consumption and circuit cost increase
Solution Approach 1:
The patent changes the operating parameters by implementing dynamic frequency scaling where the circuit operates at different frequencies depending on the workload. During time slots when processing a single bitstream, the circuit operates at lower frequency, and only increases frequency when multiple bitstreams require simultaneous processing, thereby reducing average power consumption while maintaining required processing speed
Solution Approach 2:
The patent applies partial action by implementing selective motion compensation where not all macroblocks require full motion compensation processing. The system identifies regions with significant motion and applies motion compensation only to those regions, while stationary regions use simpler processing, thereby reducing the total processing time and allowing lower operating frequency
Data Source
AI summary
An image processing device and an image processing method are used for processing a video data that includes a first bitstream and a second bitstream. The image processing method includes the following steps: decoding the first bitstream to generate a first frame; generating an intermediate image based on at least the first frame; and mixing the intermediate image and a buffered image according to a reference data to generate a pre-mixed image. The buffered image includes a second frame and a third frame which correspond to the second bitstream and the first bitstream, respectively.