Research on the Automatic Recognition Method of Micro-nano Regeneration Rubber Filler Dispersion in Scanning Electron Microscope Images

Micro-nano regenerative rubber represents a significant advancement over traditional regenerative rubber by eliminating the need for polluting chemical agents such as coal tar, asphalt, and regenerators. The performance of composite materials filled with micro-nano regenerative rubber depends critically on the degree of filler dispersion. However, current methods for assessing dispersion rely on subjective visual estimation of electron microscope images, which lack quantitative precision and are prone to human error, especially given the complex distribution of micro-nano particles. This study introduces a novel, automated method for grading the dispersion of micro-nano regenerative rubber fillers in scanning electron microscope (SEM) images, leveraging advanced image processing techniques. The proposed method automatically identifies and quantifies the dispersion of micro-nano regenerative rubber clusters through particle size and distribution analysis. By establishing objective grading rules for dispersion levels, it provides a reliable and efficient alternative to traditional subjective assessments. This innovative approach not only enhances the accuracy of dispersion evaluation but also facilitates the development of high-performance composite materials by enabling precise control over filler distribution. The method's effectiveness is validated through comprehensive experimental analysis, demonstrating its potential to significantly improve the quality and consistency of micro-nano regenerative rubber applications in various industries.