Automated Insect Identification through Concatenated Histograms of Local Appearance Features

Abstract

This paper describes a computer vision approach to automated rapid-throughput taxonomic identification of stonefly larvae. The long-term goal of this research is to develop a cost-effective method for environmental monitoring based on automated identification of indicator species. Recognition of stonefly larvae is challenging because they are highly articulated, they exhibit a high degree of intraspecies variation in size and color, and some species are difficult to distinguish visually, despite prominent dorsal patterning. The stoneflies are imaged via an apparatus that manipulates the specimens into the field of view of a microscope so that images are obtained under highly repeatable conditions. The images are then classified through a process that involves (a) identification of regions of interest, (b) representation of those regions as SIFT vectors [1], (c) classification of the SIFT vectors into learned "features" to form a histogram of detected features, and (d) classification of the feature histogram via state-of-the-art ensemble classification algorithms. The steps (a) to (c) compose the concatenated feature histogram (CFH) method. We apply three region detectors for part (a) above, including a newly developed principal curvature-based region (PCBR) detector. This detector finds stable regions of high curvature via a watershed segmentation algorithm. We compute a separate dictionary of learned features for each region detector, and then concatenate the histograms prior to the final classification step.Send offprint requests to: Tom Dietterich, 1145 Kelley Engineering Center, Corvallis, OR 97331-5501, USA, tgd@eecs.oregonstate.eduWe evaluate this classification methodology on a task of discriminating among four stonefly taxa, two of which, Calineuria and Doroneuria, are difficult even for experts to discriminate. The results show that the combination of all three detectors gives four-class accuracy of 82% and three-class accuracy (pooling Calineuria and Doroneuria) of 95%. Each region detector makes a valuable contribution. In particular, our new PCBR detector is able to discriminate Calineuria and Doroneuria much better than the other detectors.

A Review of Codebook Models in Patch-Based Visual Object Recognition

“…This straightaway increases the model's complexity by T × K, making it difficult to cope with a large number of images and a large number of object categories. Evaluations are made on the Stonefly image dataset [24] containing 3,826 images of nine different species. An ensemble of 50 unpruned C4.5 decision trees [49] was employed in each boosting iteration.…”

“…A random forest is associated with each and every combination of the detector and descriptor. Experiments were carried out with the Stonefly-9 [24] image dataset containing 3,826 images of nine different species, and the PASCAL VOC Challenge 2006 image dataset containing ten classes. For the PASCAL06 image set, interest points in each image were detected using Harris, Hessian and PCBR detector [10] and regularly sampled image patches.…”

Identification of Pecan Weevils through Image Processing

“…There are several insect identification systems that have been developed including: Digital Automated Identification System (DAISY) (Watson et al, 2003), Automated Bee Identification System (ABIS) (Arbuckle et al, 2001), Species Identification Automated and Web Accessible (SPIWA) (Do and Harp, 1999) and the Automated Insect Identification through Concatenated Histograms of Local Appearance (AIICHLA) (Larios et al, 2007). However, these systems have some limitations and may not be applicable for identifying pecan weevils.…”

Network of evolutionary binary classifiers for classification and retrieval in macroinvertebrate databases

“…Evidence suggests that automated recognition can match human taxa identification accuracy at greatly reduced costs [1], as in [2], [3], [4], and [5] different classifiers (Support Vector Machines, Self Organizing Maps, Artificial Neural Networks) with fixed configurations have been used for automated insect classifications. So far the development of automated identification techniques for freshwater macroinvertebrates has received very little attention [6]. In a recent study [7] on a set of river macroinvertebrates, aerage correct classification of 88.2% and 75.3% have been achieved in training and test sets.…”