To address the issues of low identification accuracy and poor efficiency in rock thin section lithology identification， a rock thin section lithology identification model based on improved GhostNet was proposed. First， an edge guidance module was constructed using multi-directional Sobel operator to extract edge and texture information from rock thin section images effectively. Second， CBAM （Convolutional Block Attention Module） was embedded at key positions within the GhostNet model to enhance the feature extraction capability and further improve the model's identification performance. Finally， the classifier structure was redesigned by replacing the original classifier with a global average pooling layer， so as to reduce the model's parameters， thereby improving the model training efficiency. A hierarchical classification method was adopted to divide the dataset into three major categories and 108 small categories for coarse-grained and fine-grained identification， respectively. Experimental results showed that the proposed model achieved a first-level classification accuracy of 98.15% and a second-level classification accuracy of 96.39% on the test set， with a model size of 16.3 MB. Compared with EfficientNetV2， the proposed model achieved a 1.31 percentage point improvement in first-level classification accuracy and an 81% reduction in parameters. The proposed model demonstrates superior classification performance and provides an efficient and accurate solution for lithology identification of rock thin sections.