Identification Of Moths Research Articles

Ultrastructure of Egg Chorion and Its Sculpture in Two Arctiids (Erebidae: Arctiinae)

The identification of moths has been always dependent on characters like wing maculation, wing venation, male and female genitalic features, but the ultrastructural characters like egg chorion patterns have been completely ignored. These are equally important for identification. In the present study, scanning electron microscope was used to inspect, characterize, and depict eggs of two tiger moth species i.e., Andala unifascia Walker, 1855 and Creatonotos transiens (Walker, 1855) referable to family Erebidae. On the eggshells of both these species, morphological features have been observed and these are of great taxonomic value at species as well as at genus levels. These characters will enable identifying moths at much earlier stages of their life histories i.e., eggs.

Molecular Advances in Larval Fruit Moth Identification to Facilitate Fruit Export From Western United States Under Systems Approaches

AbstractMolecular advances facilitate fruit export by improving rapid pest diagnosis by polymerase chain reaction (PCR) and advanced sequencing technology. Improved pest detection provides timely certification of the quarantine pest-free status in the commodity being exported, avoiding unnecessary commodity treatment. The U.S.–Japan Systems Approach to export fresh cherries from the Western United States that targets the codling moth, Cydia pomonella (Linnaeus) (Lepidoptera: Tortricidae), is used as an example. Suspect codling moth larvae interdicted at cherry packing houses are distinguished by PCR from other internal fruit moth larvae such as the oriental fruit moth, Grapholita molesta (Busck) (Lepidoptera: Tortricidae); lesser appleworm, G. prunivora (Walsh) (Lepidoptera: Tortricidae); cherry fruitworm, G. packardi (Zeller) (Lepidoptera: Tortricidae); and filbertworm, Cydia latiferreana (Walsingham) (Lepidoptera: Tortricidae). Identification is confirmed by sequencing the amplicon of a 301 bp region of the COI gene produced by PCR of the DNA from a suspect moth and comparing this sequence of COI gene sequences of other internal fruit feeders of pome fruit. This sequence comparison results in unambiguous pest identification. These findings are discussed in the context of systems approach research to meet evolving needs of phytosanitary requirements for global export of fruits.

Moth (Insecta: Lepidoptera) fauna of Sariska Tiger Reserve, Rajasthan, India

Survey to assess moth diversity was carried out in Sariska Tiger Reserve of Rajasthan in 2019 from June to November. Total 16 species were recorded under 15 genera, belonging to 5 families for the first time from Sariska Tiger Reserve. Of these, 12 species viz. Eudocima materna Linnaeus, 1767; Eudocima phalonia Linnaeus, 1763; Dysgonia stuposa (Fabricius, 1794); Euproctis divisia Walker, 1855; Achaea janata Linnaeus, 1758; Acherontia styx Westwood, 1847; Daphnis nerii Linnaeus, 1758; Hippotion rosetta (Swinhoe, 1892); Agrius convolvuli Linnaeus, 1758; Nephele hespera Fabricius, 1775; Theretra oldenlandiae (Fabricius, 1775); Caligula lindia Moore, 1865 were first time reports from the state. Moth species were collected with the help of a light sheet system operated every night from 9:00 pm to 2:00 am. Ethyl acetate fumes were used to kill collected speciemens and processed as per standard strategies in Lepidopterology. The identification of moths was carried out with the help of identification keys, standard reference keys. Family Erebidae and Sphingidae are represented by six species each, followed by Saturniidae represented by two species, Noctuidae and Crambidae represented by one species each. This study will improve our understanding of Sariska’s biodiversity and be used to develop strategies for the conservation of moth diversity.

Developing an automatic identification and early warning and monitoring web based system of fall army worm based on machine learning in developing countries

To combat the fall Army worm (FAW-Spodoptera frugiperda) pest which has a negative impact on world food security, there is need to come up with methods that can be used alongside conventional methods of spraying. Therefore this paper proposes a machine learning based system for automatic identification and monitoring of Fall Army worm Moths. The system will aim to address challenges that are associated with trap based FAW monitoring such as manual data collection as the system will automate the data collection process. The study will aim to automate the data collection process by developing a machine learning algorithm for FAW moth identification. The study will develop web and mobile applications integrated with Geographic information system (GIS) technology in addition to trap automation. The tools developed in this study will aim to improve the accuracy and efficiency of FAW monitoring by reducing the aspect of human intervention. At the time of writing this paper, only the web based tool prototype has been developed, therefore this paper mostly focuses on the design of the web based tool. The paper also provides a brief quantification of the chosen machine learning technique to be used in the study.

Machine Learning Algorithms for automated Image Capture and Identification of Fall Armyworm (FAW) Moths

Automated entomology is one of the field that has received a fair attention from the computer scientists and its support disciplines. This can further be confirmed by the recent attention that the Fall Armyworm (FAW) (Spodoptera frugiperda) has received in Africa particularly the Southern African Development Community (SADC). As the FAW is known for its devastating effects, stakeholders such as the Food and Agriculture Organization (FAO), SADC and University of Zambia (UNZA) have agreed to develop robust early monitoring and warning system. To supplement the stakeholders’ efforts, we choose a branch of artificial intelligence that employs deep neural network architectures known as Google TensorFlow. It is an advanced state-of-the-art machine learning technique that can be used to identify the FAW moths. In this paper, we use Google TensorFlow, an open source deep learning software library for defining, training and deploying machine learning models. We use the transfer learning technique to retrain the Inception v3 model in TensorFlow on the insect dataset, which reduces the training time and improve the accuracy of FAW moth identification. Our retrained model achieves a train accuracy of 57 – 60 %, cross entropy of 65 – 70% and validation accuracy of

Genetic variation within species and haplotypes of the Wiseana (Lepidoptera: Hepialidae) complex and development of non-sequenced based identification tools to aid field studies.

Porina is the common name used to describe moths and caterpillars of the seven endemic species and three haplotypes of Wiseana in New Zealand. Several species have adapted to eating introduced pasture plants; however, a paucity of defining morphological characteristics has meant that porina are grouped as indistinguishable species within a complex. This study aimed to develop non-sequencing identification methods for porina species. We sequenced 1718 bp of the mitochondrial cytochrome oxidase genes for all Wiseana species and haplotypes using 75 specimens collected from 42 sites throughout New Zealand. Two identification methods were developed; the high-resolution melt method uses 106 bp of the cytochrome oxidase I gene, and the restriction fragment polymorphism method uses a larger 700 bp region from the same gene. Validation was performed on a further 275 specimens. Melt curve profiles varied more with population heterogeneity than digest results. In both methods, W. mimica and W. jocosa were inseparable. W. fuliginea grouped with W. mimica and W. jocosa in the restriction method, but the melt curve profile differed. Until now, ecological studies of porina at the species level have been implausible. Our non-sequencing based methods allow rapid identification of moths and caterpillars to species and haplotype level, paving the way for ecological studies of pasture pest species and the development of species-specific mitigation strategies. © 2017 Society of Chemical Industry.

A software system for automated identification and retrieval of moth images based on wing attributes

Manually collecting, identifying, archiving and retrieving specimen images is an expensive and time-consuming work for entomologists. There is a clear need to introduce fast systems integrated with modern image processing and analysis algorithms to accelerate the process. In this paper, we describe the development of an automated moth species identification and retrieval system (SPIR) using computer vision and pattern recognition techniques. The core of the system is a probabilistic model that infers Semantically Related Visual (SRV) attributes from low-level visual features of moth images in the training set, where moth wings are segmented into information-rich patches from which the local features are extracted, and the SRV attributes are provided by human experts as ground-truth. For the large amount of unlabeled test images in the database or added into the database later on, an automated identification process is evoked to translate the detected salient regions of low-level visual features on the moth wings into meaningful semantic SRV attributes. We further propose a novel network analysis based approach to explore and utilize the co-occurrence patterns of SRV attributes as contextual cues to improve individual attribute detection accuracy. Working with a small set of labeled training images, the approach constructs a network with nodes representing the SRV attributes and weighted edges denoting the co-occurrence correlation. A fast modularity maximization algorithm is proposed to detect the co-occurrence patterns as communities in the network. A random walk process working on the discovered co-occurrence patterns is applied to refine the individual attribute detection results. The effectiveness of the proposed approach is evaluated in automated moth identification and attribute-based image retrieval. In addition, a novel image descriptor called SRV attribute signature is introduced to record the visual and semantic properties of an image and is used to compare image similarity. Experiments are performed on an existing entomology database to illustrate the capabilities of our proposed system. We observed that the system performance is improved by the SRV attribute representation and their co-occurrence patterns.

Pose estimation-dependent identification method for field moth images using deep learning architecture

Due to the varieties of moth poses and cluttered background, traditional methods for automated identification of on-trap moths suffer problems of incomplete feature extraction and misidentification. A novel pose estimation-dependent automated identification method using deep learning architecture is proposed in this paper for on-trap field moth sample images. To deal with cluttered background and uneven illumination, two-level automated moth segmentation was created for separating moth sample images from each trap image. Moth pose was then estimated in terms of either top view or side view. Suitable combinations of texture, colour, shape and local features were extracted for further moth description. Finally, the improved pyramidal stacked de-noising auto-encoder (IpSDAE) architecture was proposed to build a deep neural network for moth identification. The experimental results on 762 field moth samples by 10-fold cross-validation achieved a good identification accuracy of 96.9%, and indicated that the deployment of the proposed pose estimation process is effective for automated moth identification.