Grape Maturity Estimation Using Time-of-Flight and LiDAR Depth Cameras

Area of Science:

• Agricultural Technology
• Computer Vision
• Horticulture

Background:

• Accurate postharvest maturity assessment is crucial for table grape quality.
• Traditional methods can be destructive or labor-intensive.
• Non-contact, non-destructive methods are needed for efficient quality control.

Purpose of the Study:

• To evaluate the feasibility of using low-cost depth cameras for estimating green table grape maturity after harvest.
• To investigate the relationship between depth scan data and grape aging.

Main Methods:

• Utilized time-of-flight (Kinect Azure) and LiDAR (Intel L515) depth cameras to capture berry scans.
• Analyzed depth scan data for changes in shape and distance bias over time.
• Quantified the correlation between observed distance bias and grape maturity.

Main Results:

• Depth scans showed increased distance bias as grapes aged due to light scattering.
• The distance bias variation correlated well with time, with R2 values of 0.969 (Kinect Azure) and 0.904 (Intel L515).
• A flattened peak in the depth scan data indicated increased grape maturity.

Conclusions:

• Time-of-flight and LiDAR cameras show significant potential for non-contact, non-destructive grape maturity estimation.
• Depth camera technology offers a promising avenue for postharvest quality assessment in viticulture.
• This method could lead to improved quality control and reduced waste in the grape industry.