Towards generic image classification using tree-based learning: An extensive empirical study

Abstract

This paper considers the general problem of image classification without using any prior knowledge about image classes. We study variants of a method based on supervised learning whose common steps are the extraction of random subwindows described by raw pixel intensity values and the use of ensemble of extremely randomized trees to directly classify images or to learn image features. The influence of method parameters and variants is thoroughly evaluated so as to provide baselines and guidelines for future studies. Detailed results are provided on 80 publicly available datasets that depict very diverse types of images (more than 3800 image classes and over 1.5 million images).

The Need for Careful Data Collection for Pattern Recognition in Digital Pathology

“…In this note, we summarized data collection challenges in this field and suggest guidelines and tools to improve the quality of ground-truth datasets. Overall, we hope these comments will complement other recent studies that provide guidelines for the design and application of pattern recognition methodologies,[313435] hence contribute to the successful application of pattern recognition in digital pathology.…”

Using a Structural Root System Model to Evaluate and Improve the Accuracy of Root Image Analysis Pipelines

“…Random Forests can perform non-linear predictions and, thus, often outperform linear models. Since its introduction by Breiman (2001), Random Forests have been widely used in many fields from gene regulatory network inference to generic image classification (Huynh-Thu et al, 2013; Marée et al, 2016). Random Forest relies on growing a multitude of decision trees, a prediction algorithm that has shown good performances by itself but, when combined with other decision trees (hence the name forest), returns predictions that are much more robust to outliers and noisy data (see bootstrap aggregating, Breiman, 1996).…”

An approach for detection of glomeruli in multisite digital pathology

“…In this work, we propose to evaluate two generic classifiers readily available, namely WND-CHARM [15] and the ET-FL (Extremely Randomized Trees for Feature Learning) tree-based method of [11]. These methods have been shown effective on various classification tasks.…”

“…We also evaluate systematically the ET-FL classification method of [11] based on random subwindows and extremely randomized tree learning method. This method first performs random extraction of a large number of square subwindows in candidate images then it uses trees to build a novel image description (a global, sparse, feature vector that encodes subwindow frequencies in tree leaves) subsequently classified by a linear SVM classifier.…”

“…This method first performs random extraction of a large number of square subwindows in candidate images then it uses trees to build a novel image description (a global, sparse, feature vector that encodes subwindow frequencies in tree leaves) subsequently classified by a linear SVM classifier. For each of the normalization variant presented previously, we trained models using different configurations of subwindow size intervals (as in [11]), and we report the best obtained result. Default tests were made using 100 training subwindows per candidate image.…”