Design of an Integrated Near-Infrared Spectroscopy Module for Sugar Content Estimation of Apples

Area of Science:

• Agricultural Engineering
• Spectroscopy
• Embedded Systems

Background:

• Accurate estimation of fruit sugar content is crucial for quality control and harvest timing.
• Near-infrared (NIR) spectroscopy offers a non-destructive method for analyzing fruit composition.
• Developing compact and integrated NIR modules is essential for portable and on-site applications.

Purpose of the Study:

• To design and develop an integrated near-infrared (NIR) spectroscopy prototype module for estimating apple sugar content.
• To optimize wavelength selection and analog-to-digital converter (ADC) resolution for efficient sugar estimation.
• To evaluate the performance of the integrated NIR module in terms of accuracy and precision.

Main Methods:

• Utilized a tungsten halogen light source and a 700 nm-1000 nm NIR spectrometer with 10 nm resolution and 16b ADC.
• Determined optimal wavelengths by analyzing the correlation coefficient (R²) among 31 distributed wavelengths.
• Reduced ADC resolution iteratively while monitoring R² to find the most effective resolution.
• Designed an integrated NIR module using selected eight wavelengths and 13b ADC resolution, implemented in a 0.18 μm CMOS process.

Main Results:

• The integrated NIR module, occupying 0.84 mm², was implemented using a 0.18 μm 1P6M CMOS process.
• The module utilized eight off-chip light-emitting diodes (LEDs) and one photodiode (PD).
• The measured correlation coefficient (R²) was 0.365, with a standard error of calibration (SEC) of 0.686 brix.

Conclusions:

• An integrated NIR spectroscopy module for apple sugar content estimation was successfully designed and fabricated.
• The optimized design achieved a moderate R² of 0.365, indicating the feasibility of the approach.
• Further research may focus on improving the R² and SEC for more precise sugar content estimation in apples.