Use iPhone 17 LiDAR to measure plant leaf curvature and growth rates

LiDAR works by emitting pulses of laser light towards the target object—here, plant leaves—and measuring the time it takes for the light to reflect back to the sensor. This time-of-flight data allows the device to construct 3D spatial maps of the leaf surface with millimeter-level accuracy. The iPhone 17 integrates this sensor with advanced processing hardware and software, capable of producing dense point clouds combined with RGB data to capture both geometrical and color information essential for plant studies.

Measuring Leaf Curvature

To quantify leaf curvature using the iPhone 17 LiDAR, a typical workflow involves scanning the leaf surface from multiple angles, often employing a semicircular movement around the leaf to ensure full coverage of its upper and lower surfaces. The 3D point cloud generated from these scans can be processed using mesh reconstruction algorithms, such as the Alpha Shapes method or Poisson surface reconstruction, which create continuous surface models from discrete points.

Once a 3D surface mesh is created, leaf curvature is derived by analyzing local surface geometry. Curvature can be quantified as either Gaussian curvature, which represents intrinsic curvature, or mean curvature related to bending. By calculating curvature across the entire leaf surface, variations in leaf folding, rolling, or other deformation patterns can be objectively characterized. This facilitates studies on leaf mechanics, stress responses, and adaptation mechanisms.

Growth Rate Estimation

Tracking plant leaf growth rates with iPhone 17 LiDAR involves repeated scans of the same leaf or leaves over defined time intervals. Careful registration of 3D point clouds collected at different times allows precise detection of changes in leaf size, shape, and volume. Growth metrics can be extracted by comparing surface area, leaf length, and even curvature changes from successive scans.

Automated segmentation algorithms can isolate individual leaves from complex plant architectures within the 3D point cloud, enabling per-leaf growth analyses without destructive sampling. Leveraging machine learning or image processing techniques helps automate extraction of morphometric parameters such as length, width, surface area, and curvature.

Practical Implementation and Accuracy

Although first demonstrated effectively on earlier iPhone models like the iPhone 13 Pro, recent improvements in the iPhone 17's LiDAR technology have further enhanced spatial resolution, point density, and scanning speed. Field experiments scanning maize and fruit tree leaves illustrate that LiDAR data from iPhones deliver strong correlations with traditional measurement tools (such as area meters and manual calipers), with R-squared values commonly exceeding 0.85 for surface area and morphological traits.

To maximize accuracy, environmental conditions and scanning protocols must be well controlled: consistent lighting, minimal plant motion, and careful sensor-path control during scanning all improve data quality. The iPhone 17's built-in gyroscope, magnetometer, and advanced processor assist in stabilizing scans and compensating for movement.

Software and Data Processing Tools

Specialized applications like Polycam or custom-developed software leveraging the iPhone 17's LiDAR output allow users to generate 3D point clouds and process them for plant morphometrics. These tools enable:

- Dense point cloud capture with RGB color attributes.
- 3D reconstruction through mesh generation.
- Segmentation of leaf structures from the entire plant model.
- Calculation of leaf surface area, curvature, and volumetric traits.
- Temporal analysis by aligning repeated scans for growth tracking.

Programming libraries such as Open3D provide algorithms for surface reconstruction (Alpha Shapes, Poisson), mesh analysis for curvature, and point cloud registration needed for temporal comparison. Python-based toolchains can automate processing pipelines from raw LiDAR data to usable growth metrics.

Applications in Plant Science

Using LiDAR on smartphones like the iPhone 17 brings high-precision, rapid phenotyping directly to the field or greenhouse. It facilitates:

- Non-destructive measurements of leaf morphology and curvature, reflecting environmental stress or genetic traits.
- Continuous monitoring of growth rates at fine spatial and temporal scales without physically disturbing plants.
- Enhanced understanding of physiological parameters related to photosynthesis, transpiration, and overall plant health linked to leaf surface geometry.

This approach is cost-effective and scalable compared to traditional laser scanners, enabling widespread adoption in precision agriculture, forestry research, and plant breeding programs.

Summary of the Process

1. Setup the plant specimen in a stable position with minimal wind.
2. Use the iPhone 17 LiDAR, moving it semicircularly and around the plant for full 3D coverage.
3. Capture point cloud data enriched with RGB using apps that interface with the LiDAR sensor.
4. Process the 3D point cloud with reconstruction algorithms to create accurate leaf surface meshes.
5. Analyze the meshes for curvature by calculating local surface geometrical properties.
6. Repeat scans at time intervals to quantify growth rates by measuring changes in leaf size and surface traits.
7. Use automated or semi-automated segmentation to isolate individual leaves for specific analysis.
8. Validate measurements with traditional methods or known standards to ensure accuracy.