CMOS imager non-uniformity response correction-based high-accuracy spot target localization.

Area of Science:

• Image sensor technology
• Optical metrology
• Scientific instrumentation

Background:

• Accurate pixel response in complementary metal-oxide semiconductor (CMOS) image sensors is critical for high-precision applications such as astronomy and medical imaging.
• Pixel non-uniformity, arising from variations in silicon structure and readout electronics, necessitates effective flat-field correction before image processing.
• Traditional full-plane calibration methods may not always yield optimal results for specific high-accuracy detection tasks.

Purpose of the Study:

• To propose a new flat-field model and correction method for CMOS image sensors.
• To enhance the accuracy of pixel response, particularly in spot target imaging areas.
• To improve the signal-to-noise ratio (SNR) for high-accuracy spot target localization.

Main Methods:

• Developed a flat-field correction model focusing on spot scale areas of CMOS sensor pixel response.
• Implemented a method to fit the mean response curve of selected normal pixels across varying light intensities and exposure times.
• Evaluated the correction method's impact on spot target extraction accuracy.

Main Results:

• Reduced non-uniform variance from 7.34 (LSB/10 bit) to 1.91 (LSB/10 bit), an improvement of 74.1%.
• Significantly decreased the noise effect on spot extraction accuracy, improving it from 0.3453 pixels to 0.0116 pixels (1σ).
• Demonstrated enhanced imaging SNR for high-accuracy spot target localization.

Conclusions:

• The proposed spot-area-based flat-field correction method effectively addresses non-uniform pixel response in CMOS image sensors.
• This approach guarantees higher accuracy in spot imaging areas, crucial for demanding scientific and medical applications.
• The method provides a substantial improvement in spot target extraction accuracy and overall image quality.