VLSI Design Based on Block Truncation Coding for Real-Time Color Image Compression for IoT.

Area of Science:

• Electrical Engineering
• Computer Engineering
• Medical Imaging Technology

Background:

• Real-time patient data acquisition in emergencies is critical for hospitals.
• Existing image compression methods often lack the efficiency required for IoT devices in critical care.

Purpose of the Study:

• To develop a high-compression-ratio, real-time image compression VLSI design for IoT image sensors.
• To improve the efficiency of image data transmission from medical devices in emergency situations.

Main Methods:

• A novel algorithm combining YEF transform, color sampling, block truncation coding (BTC), threshold optimization, sub-sampling, prediction, quantization, and Golomb-Rice coding.
• Machine learning was employed to train optimal BTC parameters.
• A low-complexity, pipelined VLSI architecture was designed using TSMC 0.18 μm CMOS technology.

Main Results:

• The proposed algorithm achieves a higher compression ratio compared to traditional BTC methods.
• The VLSI chip operates at 100 MHz, with a core area of 598,880 μm² and 56.3 K logic gates.
• The design supports real-time processing at 50 frames per second, suitable for CMOS image sensors.

Conclusions:

• The developed image compression design offers significant improvements in compression ratio and real-time performance.
• This technology is well-suited for real-time CMOS image sensor applications, particularly in emergency medical scenarios.