Rapid and reliable identification of insects is important in many contexts, from the detection of disease vectors and invasive species to the sorting of material from biodiversity inventories. Because of the shortage of adequate expertise, there has long been an interest in developing automated systems for this task. Previous attempts have been based on laborious and complex handcrafted extraction of image features, but in recent years it has been shown that sophisticated convolutional neural networks (CNNs) can learn to extract relevant features automatically, without human intervention. Unfortunately, reaching expert-level accuracy in CNN identifications requires substantial computational power and huge training data sets, which are often not available for taxonomic tasks. This can be addressed using feature transfer: a CNN that has been pretrained on a generic image classification task is exposed to the taxonomic images of interest, and information about its perception of those images is used in training a simpler, dedicated identification system. Here, we develop an effective method of CNN feature transfer, which achieves expert-level accuracy in taxonomic identification of insects with training sets of 100 images or less per category, depending on the nature of data set. Specifically, we extract rich representations of intermediate to high-level image features from the CNN architecture VGG16 pretrained on the ImageNet data set. This information is submitted to a linear support vector machine classifier, which is trained on the target problem. We tested the performance of our approach on two types of challenging taxonomic tasks: 1) identifying insects to higher groups when they are likely to belong to subgroups that have not been seen previously and 2) identifying visually similar species that are difficult to separate even for experts. For the first task, our approach reached $CDATA[$CDATA[$>$$92% accuracy on one data set (884 face images of 11 families of Diptera, all specimens representing unique species), and $CDATA[$CDATA[$>$$96% accuracy on another (2936 dorsal habitus images of 14 families of Coleoptera, over 90% of specimens belonging to unique species). For the second task, our approach outperformed a leading taxonomic expert on one data set (339 images of three species of the Coleoptera genus Oxythyrea; 97% accuracy), and both humans and traditional automated identification systems on another data set (3845 images of nine species of Plecoptera larvae; 98.6 % accuracy). Reanalyzing several biological image identification tasks studied in the recent literature, we show that our approach is broadly applicable and provides significant improvements over previous methods, whether based on dedicated CNNs, CNN feature transfer, or more traditional techniques. Thus, our method, which is easy to apply, can be highly successful in developing automated taxonomic identification systems even when training data sets are small and computational budgets limited. We conclude by briefly discussing some promising CNN-based research directions in morphological systematics opened up by the success of these techniques in providing accurate diagnostic tools.
- MeSH
- fylogeneze MeSH
- hmyz klasifikace MeSH
- klasifikace metody MeSH
- neuronové sítě * MeSH
- reprodukovatelnost výsledků MeSH
- zvířata MeSH
- Check Tag
- zvířata MeSH
- Publikační typ
- časopisecké články MeSH
- hodnotící studie MeSH
- práce podpořená grantem MeSH
The genus Oxythyrea Mulsant, 1842 comprises ten currently recognized species. So far, larval descriptions for only three species are available. Here, we present comprehensive descriptions of third instar larval morphology for the six previously unstudied Oxythyrea species along with detailed re-descriptions of larval morphology for the previously described species. All descriptions are supplemented by photographs and line art drawings of the morphological structures. This work also contains a key to third instar larvae of the genus Oxythyrea as well as observations from collecting sites with focus on biological characteristics of all covered species including pictures of biotopes and host plants for adults. In addition, notes on the breeding circumstances for species are provided.
The development of modern methods of species delimitation, unified under the "integrated taxonomy" approach, allows a critical examination and re-evaluation of complex taxonomic groups. The rose chafer Protaetia (Potosia) cuprea is a highly polymorphic species group with a large distribution range. Despite its overall commonness, its taxonomy is unclear and subject to conflicting hypotheses, most of which largely fail to account for its evolutionary history. Based on the sequences of two mitochondrial markers from 65 individuals collected across the species range, and a detailed analysis of morphological characters including a geometric morphometry approach, we infer the evolutionary history and phylogeography of the P. cuprea species complex. Our results demonstrate the existence of three separate lineages in the Western Palearctic region, presumably with a species status. However, these lineages are in conflict with current taxonomic concepts. None of the 29 analyzed morphological characters commonly used in the taxonomy of this group proved to be unambiguously species- or subspecies- specific. The geometric morphometry analysis reveals a large overlap in the shape of the analyzed structures (pronotum, meso-metaventral projection, elytra and aedeagus), failing to identify either the genetically detected clades or the classical species entities. Our results question the monophyly of P. cuprea in regard to P. cuprina, as well as the species status of P. metallica. On the other hand, we found support for the species status of the Sicilian P. hypocrita. Collectively, our findings provide a new and original insight into the taxonomy and phylogeny of the P. cuprea species complex. At the same time, the results represent the first attempt to elucidate the phylogeography of these polymorphic beetles.
Colilodion colongi sp. nov. of the clavigerite ant-loving beetle from Mindanao, the Philippines is described, illustrated, and distinguished from related species. The distribution map for all known species is given.
Anew hydrophilid genus Chimaerocyon gen. nov. containing two species, C. shimadai sp. nov. (Malaysia: Pahang) and C. sumatranus sp. nov. (Indonesia: Sumatra), is described. Specimens of C. shimadai were collected from brood cells in anest of Pheidole singaporensis Özdikmen, 2010. The biology of C. sumatranus remains unknown. A molecular phylogeny based on four genes (cox1, cox2, 18S and 28S) supports the placement of the genus as deeply nested within the Cercyon-group of the tribe Megasternini. This position is supported by the subdistal position of the median spur in the hind wing (unique to Megasternini) and the presence of sucking disc on male maxilla (unique for Megastemini+Sphaeridiini). The remaining external morphology differs substantially from other representatives of Megasternini. The hypothesis that the aberrant morphology of Chimaerocyon gen. nov. is a consequence of myrmecophily is discussed.
