Improving the Representation of CNN Based Features by Autoencoder for a Task of Construction Material Image Classification

Deep learning based model named Convolution Neural Network (CNN) has been extensively employed by diversified applications concerned images or videos data. Because training a specific CNN model for an application task consumes enormous machine resources and need many of the training data, consequently pre-trained models of CNN have been broadly used as the transfer-learning scenario. By the scenario, features had been learned from a pre-trained model by one source task can be proficiently sent further to another specific task in a concept of knowledge transferring. As a result, a task specific can be directly employed such pre-trained features or further train more by setting the pre-trained features as a starting point. Thereby, it takes not much time and can improve the performance from many referenced works. In this work, with a task specific on construction material images classification, we investigate on the transfer learning of GoogleNet and ResNet101 that pre-trained on ImageNet dataset (source task). By applying both of the transfer-learning schemes, they reveal quite satisfied results. The best for GoogleNet, it gets 95.50 percent of the classification accuracy by fine-tuning scheme. Where, for ResNet101, the best is of 95.00 percent by using fixed feature extractor scheme. Nevertheless, after the learning based representation methods are further employed on top of the transferred features, they expose more appeal results. By Autoencoder based representation method reveals the performance can improve more than PCA (Principal Component Analysis) in all cases. Especially, when the fixed feature extractor of ResNet101 is used as the input to Autoencoder, the classified result can be improved up to 97.83%. It can be inferred, just applying Autoencoder on top of the pre-trained transferred features, the performance can be improved by we have no need to fine-tune the complex pre-trained model.

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

[2]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[3]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[4]  Lucio Soibelman,et al.  Construction site image retrieval based on material cluster recognition , 2006, Adv. Eng. Informatics.

[5]  Ke Chen,et al.  Bridging BIM and building: From a literature review to an integrated conceptual framework , 2015 .

[6]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Andrey Dimitrov,et al.  Vision-based material recognition for automated monitoring of construction progress and generating building information modeling from unordered site image collections , 2014, Adv. Eng. Informatics.

[8]  Ioannis Brilakis,et al.  Concrete Column Recognition in Images and Videos , 2010, J. Comput. Civ. Eng..

[9]  David S. Citrin,et al.  An analogy between various machine-learning techniques for detecting construction materials in digital images , 2015, KSCE Journal of Civil Engineering.

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

[11]  Changmin Kim,et al.  Classification of major construction materials in construction environments using ensemble classifiers , 2014, Adv. Eng. Informatics.

[12]  Derek Hoiem,et al.  Geometry-Informed Material Recognition , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).