图像处理技术在泥沙粒径分析中的应用

基于数字图像处理技术的泥沙粒径分析是近年发展起来的一种新方法。传统分水岭分割方法直接作用于梯度图像,受图像噪声、目标的细小边缘及纹理等因素影响,测量精度会相对降低。通过标记重构和区域合并对传统分水岭分割法进行优化,提出改进算法。基于实测数据,围绕数字图像处理相关技术,借鉴其他领域图像分析设备和方法,以数码单反相机和Image J2x图像处理软件为基础,建立一套泥沙图像粒径分析系统。采用图像法、机械筛分法和激光分析法分别对同一批沙样进行粒径测定。对测定结果进行的分析表明数字图像处理技术可较好地用于泥沙粒径分析。

Study on application of digital image processing technique to analysis of sediment particle size

The digital image processing technique-based analysis of sediment particle size is a new method developed in recent years. As the conventional watershed segmentation method directly acts on the gradient image under the impacts from the factors of image noise， exiguous edge and texture of object， its measuring accuracy is largely lowered. The watershed segmentation method is optimized herein through sign reconstruction and area merger， and then an improved algorithm is proposed. Based on the measured data and around relevant digital image processing techniques， a set of sediment imaging particle size analysis system is established on the basis of digital SLR camera and image processing software——Image J2x with the references of the image analysis equipment and methods in the other fields concerned. Whereafter， the particle size measurement is made on a same batch of sediment samples by image processing method， mechanical sieving method and laser analysis method respectively. The analysis of the measurement result shows that the digital image processing technique can be satisfactorily applied to the sediment particle size analysis.