Automated Taxonomic Identification of Insects with Expert-Level Accuracy Using Effective Feature Transfer from Convolutional Networks.

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.

Redescriptions and new host records of chewing lice of the genus Ricinus (Phthiraptera: Ricinidae) from the Neotropical Region.

Two species of the chewing louse genus Ricinus are redescribed and illustrated: Ricinus dalgleishi Nelson, 1972 from Helmitheros vermivorum (Gmelin, 1789), a new host-louse association, and Ricinus tanagraephilus Eichler, 1956 from Euphonia laniirostris d'Orbigny & Lafresnaye, 1837. Also, new host-louse associations are recorded for Ricinus vireoensis Nelson, 1972 from Vireo pallens Salvin, 1863, and for females of an unidentified species of Ricinus sp. from Corythopis delalandi (Lesson, 1831), which are described and illustrated.

New data on the taxonomy and distribution of ten Neotropical chewing lice of the genus Myrsidea (Phthiraptera: Menoponidae), including the description of a new species.

The new species Myrsidea alexanderi is described and illustrated ex Pheugopedius maculipectus (Troglodytidae) from Honduras. Redescriptions and illustrations are given for both sexes of Myrsidea chiapensis ex Calocitta formosa from Costa Rica, and the male of M. dissimilis ex Progne chalybea from Brazil. Also, seven other previously known species or subspecies of the louse genus Myrsidea are recorded and discussed from passerine birds of the Neotropical Region, as follows: Myrsidea antiqua, Myrsidea balteri, Myrsidea diffusa, Myrsidea nesomimi borealis, Myrsidea paleno, Myrsidea psittaci and Myrsidea serini. Our data increase knowledge of intraspecific morphological variability within these species, and also of their host and geographical distribution. New host-louse associations are: Agelaioides badius for M. psittaci; Basileuterus culicivorus and Myiothlypis leucoblephara for M. paleno; Mimus saturninus for M. nesomimi borealis; and Icterus dominicensis and Molothrus rufoaxillaris for Myrsidea sp.

Chewing lice of genus Ricinus (Phthiraptera, Ricinidae) deposited at the Zoological Institute of the Russian Academy of Sciences, Saint Petersburg, Russia, with description of a new species.

We revised a collection of chewing lice deposited at the Zoological Institute of the Russian Academy of Sciences, Saint Petersburg, Russia. We studied 60 slides with 107 specimens of 10 species of the genus Ricinus (De Geer, 1778). The collection includes lectotype specimens of Ricinus ivanovi Blagoveshtchensky, 1951 and of Ricinus tugarinovi Blagoveshtchensky, 1951. We registered Ricinus elongatus Olfers, 1816 ex Turdus ruficollis, R. ivanovi ex Leucosticte tephrocotis and Ricinus serratus (Durrant, 1906) ex Calandrella acutirostris and Calandrella cheleensis which were not included in Price's world checklist. New records for Russia are R. elongatus ex Turdus ruficollis; Ricinus fringillae De Geer, 1778 ex Emberiza aureola, Emberiza leucocephalos, Emberiza rustica, Passer montanus and Prunella modularis; Ricinus rubeculae De Geer, 1778 ex Erithacus rubecula and Luscinia svecica; Ricinus serratus (Durrant, 1906) ex Alauda arvensis. New records for Kyrgyzstan are R. fringillae ex E. leucocephalos and ex Fringilla coelebs. A new record for Tajikistan is R. serratus ex Calandrella acutirostris. The new species Ricinus vaderi Valan n. sp. is described with Calandra lark, Melanocorypha calandra; from Azerbaijan, as a type host.