Seminar

Rock classification provides vital information to geosciences and geological engineering practices. However, the traditional way of rock classification is to a certain degree experience-based. Computer vision-based artificial intelligence (AI), especially convolutional neural network (CNN), a subbranch of AI, with proper training and validation can instantly and precisely perceive rock images similar to how human beings observe and classify rocks. In other words, by putting a rock image into a trained CNN model, the CNN will read the information from the image and predict what rock type in the image is. This study develops and implements a next-generation CNN called RockNet to classify six different rock types. Adopting different computational strategies, RockNet can classify rock types of different grain sizes with similar texture patterns/colors, which is challenging for other CNNs. As compared with the other landmark CNNs, RockNet has been demonstrated to have a much better prediction accuracy performance. The proposal and implementation of RockNet exemplify an interdisciplinary research study, and we are confident that this study will be instrumental to pave the way for further coupling AI and geosciences.

Additional information: Mr. ZHOU Yimeng, zhouym52@connect.hku.hk