Recognizing Multiple Types of Rocks Quickly and Accurately Based on Lightweight CNNs Model

The recognition and classification of rock lithology is an extremely important task of geological surveys. This paper proposes a new method for quickly identifying multiple types of rocks suitable for geological survey work field. Based on the two lightweight convolutional neural networks (CNNs), SqueezeNet and MobileNets, and combined with the transfer learning method, a rock lithology recognition model was established. The model was embedded into a smart phone for testing. This method was used to identify and classify the images of 28 kinds of rocks. Through a comprehensive comparison of the two models, the accuracy of SqueezeNet and MobileNets in the test dataset is 94.55% and 93.27%, respectively. Via the two models, the average recognition time of a single rock image is 557 and 836 milliseconds, and rock images with a recognition accuracy of over 96% accounted for 95% and 93% of the entire test dataset. Compared with the classification method based on rock thin section images, this method does not need to make rock thin sections. This paper meets the requirements of workers to quickly and accurately identify rocks in the work field, which improves the work efficiency and limits identification costs.

[1]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[2]  Sahin Emrah Amrahov,et al.  Fuzzy Rule-Based Image Segmentation technique for rock thin section images , 2012, 2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA).

[3]  Seungoh Ryu,et al.  Determining multiple length scales in rocks , 2000, Nature.

[4]  Eocen H. Barr,et al.  SELECTIVE STAINING OF K-FELDSPAR AND PLAGIOCLASE ON ROCK SLABS AND THIN SECTIONS* , 1960 .

[5]  Frank S. Reed,et al.  Preparation of rock thin sections , 1953 .

[6]  Na Li,et al.  A transfer learning method for automatic identification of sandstone microscopic images , 2017, Comput. Geosci..

[7]  D. Wettergreen,et al.  Automatic detection and classification of features of geologic interest , 2005, 2005 IEEE Aerospace Conference.

[8]  Pejman Tahmasebi,et al.  Image-based velocity estimation of rock using Convolutional Neural Networks , 2019, Neural Networks.

[9]  Kilian Q. Weinberger,et al.  Marginalized Denoising Autoencoders for Domain Adaptation , 2012, ICML.

[10]  Mark Sandler,et al.  MobileNetV2: Inverted Residuals and Linear Bottlenecks , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[11]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[12]  E. E. Baraboshkin,et al.  Deep Convolutions for In-Depth Automated Rock Typing , 2019, Comput. Geosci..

[13]  Bernhard Schölkopf,et al.  Domain Adaptation with Conditional Transferable Components , 2016, ICML.

[14]  Ivan Laptev,et al.  Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[15]  Zhenhua Tan,et al.  Heated Metal Mark Attribute Recognition Based on Compressed CNNs Model , 2019, Applied Sciences.

[16]  Weiguo Fan,et al.  A new image classification method using CNN transfer learning and web data augmentation , 2018, Expert Syst. Appl..

[17]  S. Haubrock,et al.  Petlab: New Zealand’s national rock catalogue and geoanalytical database , 2016 .

[18]  Zhang Liangpei,et al.  Mineral Features Extraction and Analysis Based on Multiresolution Segmentation of Petrographic Images , 2011 .

[19]  Wolfram Rühaak,et al.  Processing of rock core microtomography images: Using seven different machine learning algorithms , 2016, Comput. Geosci..

[20]  Guo Cha Recognition of Rock Images Based on Multiple Color Spaces , 2014 .

[21]  Harikumar Rajaguru,et al.  Logistic regression Gaussian mixture model and softmax discriminant classifier for epilepsy classification from EEG signals , 2017, 2017 International Conference on Computing Methodologies and Communication (ICCMC).

[22]  S. Buckley,et al.  Close-range hyperspectral imaging for geological field studies: workflow and methods , 2013 .

[23]  Hans Pichler,et al.  Rock-forming minerals in thin section , 1997 .

[24]  Alexander Wong,et al.  SquishedNets: Squishing SqueezeNet further for edge device scenarios via deep evolutionary synthesis , 2017, ArXiv.

[25]  Diaz D. Santika,et al.  MobileNet Convolutional Neural Networks and Support Vector Machines for Palmprint Recognition , 2019, Procedia Computer Science.

[26]  Laercio B. Goncalves,et al.  Macroscopic Rock Texture Image Classification Using an Hierarchical Neuro-Fuzzy System , 2009, 2009 16th International Conference on Systems, Signals and Image Processing.

[27]  Mariusz Mlynarczuk,et al.  The application of pattern recognition in the automatic classification of microscopic rock images , 2013, Comput. Geosci..

[28]  Bo Chen,et al.  MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications , 2017, ArXiv.

[29]  Xiong Luo,et al.  Depthwise Separable Convolution Feature Learning for Ihomogeneous Rock Image Classification , 2018, ICCSIP.

[30]  Ben Poole,et al.  Categorical Reparameterization with Gumbel-Softmax , 2016, ICLR.

[31]  Naresh Singh,et al.  Textural identification of basaltic rock mass using image processing and neural network , 2010 .

[32]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[33]  Roberto Tagliaferri,et al.  Textural identification of carbonate rocks by image processing and neural network: Methodology proposal and examples , 2005, Comput. Geosci..

[34]  Xiaochun Luo,et al.  Measuring rock surface strength based on spectrograms with deep convolutional networks , 2019, Comput. Geosci..

[35]  Yan Li,et al.  Quick Recognition of Rock Images for Mobile Applications , 2018 .

[36]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[37]  H. Massonne,et al.  SIMPLE IDENTIFICATION AND QUANTIFICATION OF MICRODIAMONDS IN ROCK THIN-SECTIONS , 1998 .

[38]  Chao Yang,et al.  A Survey on Deep Transfer Learning , 2018, ICANN.

[39]  Guojian Cheng,et al.  Rock images classification by using deep convolution neural network , 2017 .

[40]  Lei Shu,et al.  Unsupervised feature learning for autonomous rock image classification , 2017, Comput. Geosci..