Abstract:Natural coasts define the interface between land and ocean and constitute a natural barrier to protect humans and infrastructure from waves and storm surges. In the context of global sea level rise and coastal erosion, beaches, especially sand-gravel beaches, can achieve better marine energy reduction at the expense of a smaller spatial buffer zone (i.e., the lateral range of the coastal zone), meaning that nature-based coastal reconstruction and protection are becoming increasingly important. The need for beach replenishment and coastal protection necessitates fast, accurate, portable, and wide-range gravel grain size measurement technology to provide technical support and data for the designing and monitoring of artificial gravel beaches. However, existing grain size analysis methods for gravel-sized sediment are either inefficient (such as sieve analysis) or have a limited range of sizes (such as dynamic image analyzers, up to −3 φ), or have difficulty in eliminating errors caused by gravel overlap (e.g., remote-sensing in situ image analysis). In this study, a Gravel Grain Size and Shape Automatic Measurement System (GraSSAMS) was designed to analyze grain size and shape parameters of gravel-sized sediments. The mean diameters of three types of coins and some Zirconia balls measured using this method were compared with those measured by a vernier caliper. The results show that the GraSSAMS method has good accuracy and yields an acceptable relative error. The grain sizes of 67 field gravel samples were analyzed using both the GraSSAMS method and sieve analysis. The irreducible random error between the two methods is 1.41 φ. In addition, a Camsizer XT and GraSSAMS were used to measure samples with a grain size range of 1∼8 mm (0 φ ∼ -3 φ), respectively. The results of the two methods show good correspondence, with an irreducible random error of 0.31 φ. GraSSAMS was also tested in the field on several gravel beaches. Results revealed that GraSSAMS could complete the surface gravel measurement of a 0.5 m × 0.5 m quadrat within 8 min, and the analysis results avoided the fabric error involved in remote-sensing (photographic) measurement. Based on these various results, we propose that GraSSAMS can be used to analyze the grain size and shape of sand-gravel sediments in the size range of 1∼330 mm (0 φ ∼ -8.37 φ) in a laboratory as well as in the field. Compared with traditional methods, GraSSAM has the advantages of accuracy, high efficiency, and portability and can be used as an alternative for grain size analysis of gravel sediments, with broad application prospects to gravel beaches and fluvial deposits worldwide.

Automated grain sizing from uncrewed aerial vehicles imagery of a gravel‐bed river: Benchmarking of three object‐based methods

CNN for Image-Based Sediment Detection Applied to a Large Terrestrial and Airborne Dataset

Size, shape and orientation matter: fast and semi-automatic measurement of grain geometries from 3D point clouds

Convolutional Neural Networks for Image-Based Sediment Detection Applied to a Large Terrestrial and Airborne Dataset

UAV-Based Optical Granulometry as Tool for Detecting Changes in Structure of Flood Depositions

Automatic Extraction of Mountain River Surface and Width Based on Multisource High-Resolution Satellite Images

Calibration‐free image analysis method for grain‐size distribution of small natural sand samples

Mapping erosion and deposition in an agricultural landscape: Optimization of UAV image acquisition schemes for SfM-MVS

GraSSAMS: A new instrument designed for the determination of grain size and shape of sand-gravel-sized sediment

Quantifying streambed grain sizes and hydro-biogeochemistry using YOLO and photos

Metrics for the Quantification of Seeding Characteristics to Enhance Image Velocimetry Performance in Rivers

Experimental analysis of sand grain size mapping using UAV remote sensing

An investigation of image processing techniques for substrate classification based on dominant grain size using RGB images from UAV

Estimation of grain-size distributions and associated parameters from digital images of sediment

SediNet: a configurable deep learning model for mixed qualitative and quantitative optical granulometry

Automatic Features Detection in a Fluvial Environment through Machine Learning Techniques Based on UAVs Multispectral Data

Enhancing Wheat Above-Ground Biomass Estimation Using UAV RGB Images and Machine Learning: Multi-Feature Combinations, Flight Height, and Algorithm Implications

Combining Structure from Motion and Close-Range Stereo Photogrammetry to Obtain Scaled Gravel Bar DEMs

Using 3d Point Clouds Derived From Uav Rgb Imagery To Describe Vineyard 3d Macro- Structure

Embedded Yolo-Fastest V2-Based 3D Reconstruction and Size Prediction of Grain Silo-Bag

Estimation of forage biomass and vegetation cover in grasslands using UAV imagery