Plant identification using deep convolutional networks based on principal component analysis

Official URL: http://risc01.sabanciuniv.edu/record=b1615022 (Table of Contents)

Plants have substantial effects in human vitality through their different uses in agriculture, food industry, pharmacology, and climate control. The large number of herbs and plant species and shortage of skilled botanists have increased the need for automated plant identification systems in recent years. As one of the challenging problems in object recognition, automatic plant identification aims to assign the plant in an image to a known taxon or species using machine learning and computer vision algorithms. However, this problem is challenging due to the inter-class similarities within a plant family and large intra-class variations in background, occlusion, pose, color, and illumination. In this thesis, we propose an automatic plant identification system based on deep convolutional networks. This system uses a simple baseline and applies principal component analysis (PCA) to patches of images to learn the network weights in an unsupervised learning approach. After multi-stage PCA filter banks are learned, a simple binary hashing is applied to output maps and the obtained maps are subsampled through max-pooling. Finally, the spatial pyramid pooling is applied to the downsampled data to extract features from block histograms. A multi-class linear support vector machine is then trained to classify the different species. The system performance is evaluated on the plant identification datasets of LifeCLEF 2014 in terms of classification accuracy, inverse rank score, and robustness against pose (translation, scaling, and rotation) and illumination variations. A comparison of our results with those of the top systems submitted to LifeCLEF 2014 campaign reveals that our proposed system would have achieved the second place in the categories of Entire, Branch, Fruit, Leaf, Scanned Leaf, and Stem, and the third place in the Flower category while having a simpler architecture and lower computational complexity than the winner system(s). We achieved the best accuracy in scanned leaves where we obtained an inverse rank score of 0.6157 and a classification accuracy of 68.25%.