Answers are grounded in this paper and cite their sources.
Assesses hardness classification using LIBS and machine learning models, suggesting effective detection methods.
With the development of modern industry, the requirements for metallic materials are becoming increasingly stringent. Hardness, as one of the core indicators for evaluating material properties, holds significant importance in industrial applications. Therefore, rapid and real‐time hardness detection should be carried out to further eliminate unqualified products. This study combines laser‐induced breakdown spectroscopy (LIBS) and classification models. The method identifies metals of different hardness levels quickly. The investigated samples were divided into low, medium, medium‐high, and high hardness groups, and multiple classification models were selected for comparison. Competitive adaptive reweighted sampling (CARS) was employed for feature extraction, and the extracted features were used as input data for model construction. Among the models, the convolutional neural network (CNN) achieved the best performance, with accuracy and F1‐score reaching 0.99. The results demonstrate that the combination of LIBS and classification models can be effectively applied to the detection and classification of metal hardness. Finally, SHapley Additive exPlanations (SHAP) was adopted to further analyze the contribution of features. The top‐10 elements in terms of contribution also validated the effectiveness of feature extraction and the classification model.
No takes yet. Share an insight, caveat, or question.
Che et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: