Fish Classification Research Articles

Head and neck paragangliomas: Ten years of experience in a third health center. A cohort study

Head and neck paragangliomas are rare vascular tumors derived from the paraganglionic system, located at the carotid body, jugular vein, tympanic cavity and vagal nerve.From 2010 to 2020, a cohort of 26 patients divided in two groups, 15 with cervical paragangliomas and 11 with temporal bone paragangliomas, was reviewed by analysing the medical history, the epidemiological and clinical parameters, the imaging results and classification, the modality of treatment and outcome.Cervical paragangliomas present as firm and pulsatile mass with the characteristic aspect of “salt and pepper” on MRI T1 weighted sequences. The most common type on Shamblin classification was the type II. Total surgical resection was performed in 93,33% of cases. The sensitivity of MRI in the diagnosis of vagal paragangliomas was up to 75%, with a specificity of 90,91% and the correlation of the MRI results and the findings of surgical exploration is significant with p ⩽ 0.02.Temporal bone paragangliomas appear as pulsatile mass behind the tympanic membrane, causing variable hearing loss in 90,90% of the cases. The facial nerve is the most frequently affected cranial nerve, in 36,36% of the cases. The main type according to FISH classification is the type B. Embolization was performed in all type C tumors. Surgery was the first line treatment while the inoperable patients were considered for radiotherapy.The aim of this study is to report the main clinical features of head and neck paragangliomas, the imaging tools and findings evaluating their sensitivity and specificity and the treatment protocol and outcome.

Otolith chemical fingerprints of skipjack tuna (Katsuwonus pelamis) in the Indian Ocean: First insights into stock structure delineation.

The chemical composition of otoliths (earbones) can provide valuable information about stock structure and connectivity patterns among marine fish. For that, chemical signatures must be sufficiently distinct to allow accurate classification of an unknown fish to their area of origin. Here we have examined the suitability of otolith microchemistry as a tool to better understand the spatial dynamics of skipjack tuna (Katsuwonus pelamis), a highly valuable commercial species for which uncertainties remain regarding its stock structure in the Indian Ocean. For this aim, we have compared the early life otolith chemical composition of young-of-the-year (<6 months) skipjack tuna captured from the three main nursery areas of the equatorial Indian Ocean (West, Central and East). Elemental (Li:Ca, Sr:Ca, Ba:Ca, Mg:Ca and Mn:Ca) and stable isotopic (δ13C, δ18O) signatures were used, from individuals captured in 2018 and 2019. Otolith Sr:Ca, Ba:Ca, Mg:Ca and δ18O significantly differed among fish from different nurseries, but, in general, the chemical signatures of the three nursery areas largely overlapped. Multivariate analyses of otolith chemical signatures revealed low geographic separation among Central and Eastern nurseries, achieving a maximum overall random forest cross validated classification success of 51%. Cohort effect on otolith trace element signatures was also detected, indicating that variations in chemical signatures associated with seasonal changes in oceanographic conditions must be well understood, particularly for species with several reproductive peaks throughout the year. Otolith microchemistry in conjunction with other techniques (e.g., genetics, particle tracking) should be further investigated to resolve skipjack stock structure, which will ultimately contribute to the sustainable management of this stock in the Indian Ocean.

Updated checklist of Azores Chondrichthyes (Vertebrata: Gnathostomata).

Several lists of marine fish from Azores have been published in the past. Most of those publications are difficult to access on line and several were not published in peer-reviewed journals. This checklist updates all the chondrichthyan records for the Azores Exclusive Economic Zone (EEZ), according to the most recent taxonomic classification of cartilaginous fish, as well as providing information on the conservation status for all species. We also present recent literature data on rare species and recent records for Azores. This is the first comprehensive list of cartilaginous fishes from Azores to be published in the GBIF database.

Content-based Fish Classification Using Combination of Machine Learning Methods

Fish species recognition is an increasing demand to the field of fish ecology, fishing industry sector, fisheries survey applications, and other related concerns. Traditionally, concept-based fish specifies identification procedure is used. But it has some limitations. Content-based classification overcomes these problems. In this paper, a contentbased fish recognition system based on the fusion of local features and global feature is proposed. For local features extraction from fish image, Local Binary Pattern (LBP), Speeded-Up Robust Feature (SURF), and Scale Invariant Feature Transform (SIFT) are used. To extract global feature from fish image, Color Coherence Vector (CCV) is used. Five popular machine learning models such as: Decision Tree, k-Nearest Neighbor (k-NN), Support Vector Machines (SVM), Naïve Bayes, and Artificial Neural Network (ANN) are used for fish species prediction. Finally, prediction decisions of the above machine learning models are combined to select the final fish class based on majority vote. The experiment is performed on a subset of ‘QUT_fish_data’ dataset containing 256 fish images of 21 classes and the result (accuracy 98.46%) shows that though the proposed method does not outperform all existing fish classification methods but it outperforms many existing methods and so, the method is a competitive alternative in this field.

Naive Bayesian fusion based deep learning networks for multisegmented classification of fishes in aquaculture industries

Fish classification is an essential requirement for biomass estimation, disease identification, and quality analysis. In aquaculture industries, fish classification is carried out in the processing unit. The fishes are out of water, subjecting them to structural deformation and orientation misalignments, makes classification challenging. A multisegmented fish classification technique using deep learning networks with naive Bayesian type fusion is proposed in this work to address these challenges. Fish images are acquired using an overhead camera. The fish head is identified by observing a minimal convexity deficiency region to facilitate segmentation. A multi-stage exhaustive enumerative optimization method is used to adjust the orientation, which can minimize unwanted background region in the image segment. Fish head, scales, and body are segmented from the fish image. For each fish segment, AlexNet is trained by using the transfer learning approach. A naive Bayesian fusion layer is introduced to fuse these trained deep learning networks and enhance classification accuracy. Experimental results illustrate a classification accuracy of 98.64% for ‘Fish-Pak’ image dataset with six different fish species and 98.94% for BYU fish dataset with four species. Comparative analysis with standard networks and ablation study demonstrates the accuracy and robustness of the proposed fusion architecture, respectively. Various fusion layers have also been analyzed, and observations illustrate the accuracy of the proposed NBC layer. Significant improvements in other classification performance metrics were also observed.

BettaNet: A Deep Learning Architecture for Classification of Wild Siamese Betta Species

Fish classification is a mix of animal sciences and artificial intelligence. With the advent of machine learning in artificial intelligence, classification has been done using computer vision algorithms and now deep learning is gaining prominence. Betta fish classification is not much explored. The wild species of Betta Splendens which are native to the Kingdom of Thailand are taken in the research reported in this paper. BettaNet architecture, a modified version of ResNet 152 is used to classify 6 species of wild species of betta. The experimental results show that the proposed BettaNet architecture holds better in performance in terms of accuracy and F1-scores. Two different datasets were used and the performance obtained by the proposed architecture reduced the cross-entropy loss over different experimental configurations.

インドネシア・バンガイ諸島サマ人の環境認識 --外洋漁撈をめぐる魚類・漁場・目標物の民俗分類--

インドネシア・バンガイ諸島サマ人の環境認識 --外洋漁撈をめぐる魚類・漁場・目標物の民俗分類--

Tuna classification using super learner ensemble of region-based CNN-grouped 2D-LBP models

Tuna is superior among the marine fishes that are exported in the forms of raw fish and processed food. Separation of Tuna into their species is done in industries manually, and the process is tiresome. This work proposes an automated system for classifying Tuna species based on their images. An ensemble of region-based deep neural networks is used. A sub region contrast stretching operation is applied to enhance the images. Each fish image is then divided into three regions and is augmented before giving as input to pre-trained convolutional neural networks (CNN). After fine-tuning the models, the output from the last convolutional layer is given to a grouped 2D-local binary pattern descriptor (G2DLBP). Statistical features from the descriptor are applied to different classifiers, and the best classifier for each image region model is identified. Different ensemble methods are subsequently used to combine the three CNN-G2DLBP models. Among the ensemble techniques, super learner ensemble method with random forest (RF) classifier using 5-fold cross-validation shows the highest classification accuracy of 97.32%. The performance of different ensemble methods is analyzed in terms of accuracy, precision, recall, and f-score. The proposed system shows an accuracy of 93.91% when evaluated with an independent test dataset. An ensemble of region-based CNN with textural features from G2DLBP is applied for the first time for fish classification.

IMAGE PROCESSING BASED TILAPIA SORTATION SYSTEM USING NA

Tilapia has a value of export quality and is imported from America and Europe, tilapia is cultivated in freshwater, the largest tilapia producing areas are Java and Bali for the export market in the Middle East, value fish with a size of 250 grams / head (4 fish / kg ) in their intact form is in great demand. According to news circulating, fish of this size in the Middle East are ordered to meet the consumption of workers from Asia. the fish classification process is a very difficult process to find the quality value of the fish to be sold to meet export quality. Fish classification techniques can use the GLCM technique (Gray Level Oc-Currance Matrix) classification using images of fish critters with the GLCM method.The fish image data is analyzed based on the value of Attribute, Energy, Homogenity, Correlation, Contrash, from the attribute the density data matrix is ??generated for each. Fish image data and displayed in the form of a histogram, the data from the GLCM results are then classified with the Naive Bayes algorithm, from the results of the classification of data taken from 3 types of tilapia from the types of gift, Red, and Blue.

Comparative analysis on otolith shape of &amp;lt;italic&amp;gt;Champsocephalus gunnari&amp;lt;/italic&amp;gt; and &amp;lt;italic&amp;gt;Cryodraco antarcticus&amp;lt;/italic&amp;gt; based on four morphological techniques

To study the effectiveness of otolith morphological techniques on the population classification of Antarctic ice fish, based on the collected samples of <italic>Champsocephalus gunnari</italic> and <italic>Cryodraco antarcticus</italic>, 5 kinds of basic otolith morphological parameters were measured and converted into 7 morphological indicators, the differences in otolith morphology of two species were compared and analyzed, principal component analysis (PCA), stepwise discriminant analysis (SDA) and random forest (RF) were then used to discriminate the 77 elliptical Fourier harmonics of otolith morphological parameters. The results showed that the otolith parameters and body length of two species were significantly correlated (<italic>P</italic>>0.05). The body length of <italic>C</italic>. <italic>gunnari</italic> showed a power function relationship with some morphological parameters. While otolith length, otolith height, otolith area, otolith perimeter, otolith mass of <italic>C</italic>. <italic>antarcticus</italic> were all in logarithmic relationship with the body length. It could be seen that the otolith ring rate of <italic>C</italic>. <italic>gunnari</italic> was lower than that of <italic>C</italic>. <italic>antarcticus</italic>, compared to the otolith of <italic>C</italic>. <italic>antarcticus</italic>, the otolith of <italic>C</italic>. <italic>gunnari</italic> was rounder and more regular. The conventional morphological analysis could not discriminate the otoliths of two species. The PCA could discriminate the otoliths of two species, and the discriminant success rates of SDA and RF were 98.78% and 91.67%, respectively. SDA has its advantage in small sample size, but RF is suitable for large sample size.

A mini review on characteristic of depik fish (Rasbora tawarensis) and its processed products

Depik fish (Rasbora tawarensis) is a tpical endemic freshwater fish of Laut Tawar Lake in Central Aceh, Aceh Province. This lake has become the pride of the Gayo community as a source of clean water for various needs, a source of livelihood for residents and a tourist destination. Laut Tawar Lake and depik fish have become the trademark of the city of Takengon, Central Aceh. However, it is unfortunate that due to several reasons such as environmental degradation, introduction of foreign fish, destructive fishing techniques, pollution and climate change globally, depik fish population is currently decreasing drastically. Even depik fish has been appointed as a fish with a threatened status. Studies on the existence of depik fish as local food, form of presentation and nutrition are still rarely carried out. This article tries to review the existence of depik fish and its processed products through several existing studies including classification, distribution, morphological and genetic characteristics of depik fish as well as processing and characteristics of processed products.

Monitoring Scotland’s transitional water fish communities under the EU Water Framework Directive

The history of monitoring transitional water fish in Scotland is briefly outlined. The requirements of the EU Water Framework Directive are explained and how this applies to the monitoring of transitional water fish communities in Scotland is described. The development of a monitoring programme for Scotland is outlined, including sampling methods and strategies. Six transitional waters were selected as representative for Scotland covering three different types of transitional water. A multi-metric tool, the Transitional Water Fish Classification Index was used to assess the ecological status of the fish communities in these waters and the operation of the different metrics and the creation of appropriate reference conditions is explained. The assessment tool was applied to survey data from 2005 to 2018, although only the more recent data fully met the tool requirements. The species composition and abundances in the respective transitional waters were compared. The fully valid surveys were all classed as of Good or High status, indicating the fish communities in all the representative transitional waters appeared to be in good ecological health. The efficacy of the different metrics is considered and some issues with Metric 2, enumerating migratory species, are discussed at length. A new multi-metric tool, the Estuarine Multi-metric Fish Index, is briefly discussed and its introduction for the assessments in Scotland is recommended.

Spontaneous hybridization and introgression between walleye (Sander vitreus) and sauger (Sander canadensis) in two large reservoirs: Insights from genotyping by sequencing.

Anthropogenic activities may facilitate undesirable hybridization and genomic introgression between fish species. Walleye (Sander vitreus) and sauger (Sander canadensis) are economically valuable freshwater species that can spontaneously hybridize in areas of sympatry. Levels of genomic introgression between walleye and sauger may be increased by modifications to waterbodies (e.g., reservoir development) and inadvertent propagation of hybrids in stocking programs. We used genotyping by sequencing (GBS) to examine 217 fish from two large reservoirs with mixed populations of walleye and sauger in Saskatchewan, Canada (Lake Diefenbaker, Tobin Lake). Analyses with 20,038 (r90) and 478 (r100) single nucleotide polymorphisms clearly resolved walleye and sauger, and classified hybrids with high confidence. F1, F2, and multigeneration hybrids were detected in Lake Diefenbaker, indicating potentially high levels of genomic introgression. In contrast, only F1 hybrids were detected in Tobin Lake. Field classification of fish was unreliable; 7% of fish were misidentified based on broad species categories. Important for activities such as brood stock selection, 12 of 173 (7%) fish field identified as pure walleye, and one of 24 (4%) identified as pure sauger were actually hybrids. In addition, two of 15 (13%) field‐identified hybrids were actually pure walleye or sauger. We conclude that hybridization and introgression are occurring in Saskatchewan reservoirs and that caution is warranted when using these populations in stocking programs. GBS offers a powerful and flexible tool for examining hybridization without preidentification of informative loci, eliminating some of the key challenges associated with other marker types.

Fast and Accurate Fish Classification from Underwater Video using You Only Look Once

Indonesia is a maritime country and one of the largest archipelago countries in the world. Indonesian fisheries have many types of fish in stock, this causes difficulties in introducing fish species directly. This study designed a fish species classification system using the You Only Look Once (YOLO) architecture. YOLO is an object detection method using a convolutional network that will only be just once. Unlike the convolutional networks in general that spend thousands of networks to obtain an image with computing that is long enough. The architecture of this work using YOLO9000. The dataset consists of 6 classes, that is banded butterflyfish, blue tang surgeonfish, barred hamlet, black side hawkfish, Arabian Picasso triggerfish, dan black margate grunt. System testing produces an accuracy of 92%, IoS 0.75, and 2.223 FPS using Adam optimizer. The proposed system model has good accuracy and fast detection time.

Wildfish++: A Comprehensive Fish Benchmark for Multimedia Research

In this paper, we develop a large-scale vision-language fish benchmark, namely WildFish++, for comprehensive studies in multimedia research. Concretely, WildFish++ consists of 2,348 fish categories with 103,034 images in the wild, and 3,817 fish descriptions with 213,858 words. Based on these distinct characteristics, we mainly introduce four challenging research tasks on WildFish++. (1) Fine-Grained Recognition with Comparison Texts . WildFish++ naturally contains subtle difference among fish categories, which leads to fine-grained classification. Most approaches resort to tackle this problem by capturing discriminative regions in the view of each image. However, this paradigm may be still far way from extracting the most distinct features when the context on visual difference is not available. In this case, we propose to introduce comparison fish descriptions, a unique corpus that can directly point out subtle difference between highly-confused species and naturally serve as a kind of valuable context information. With such texts, we creatively elaborate a multi-modal fish network, aiming at incorporating those comparison textual information as prior knowledge and consequently leveraging it to guide CNNs to find subtle yet distinct regions in the context of comparison texts. (2) Open-Set Classification . We often confront with unknown categories in practice, e.g., there may still exist unknown fishes in our planet. Hence, we creatively adapt WildFish++ for a novel open-set classification task, which aims at correctly assigning each test image into the unknown class or one of known classes. More importantly, we investigate a number of practical designs to boost accuracy of deep learning models in open-set scenarios. (3) Cross-Modal Retrieval . WildFish++ not only contains diversified fish images in the wild but also has rich fish descriptions about morphology diagnosis, biology information, etc. Hence, we design a challenging cross-modal retrieval task, which leverages three subtasks such as text-to-text, text-to-image, image-to-text retrieval in a unified end-to-end framework. (4) Automatic Fish Classification . Automatic fish classification is a long-term research in marine biology, while current studies are unsatisfactory due to the lack of large-scale data. In this case, we train a number of CNNs with WildFish++, and use its pre-trained models to boost fish classification on most existing benchmarks of wild fishes. We will release WildFish++ with codes/protocols ( https://github.com/PeiqinZhuang/WildFish++ ). We believe it can promote relevant studies in multimedia and beyond.

A Novel Approach to Coral Fish Detection And Classification in Underwater Footage Based on Convolutional Neural Network

Tracking and identifying fish species is crucial to understanding marine ecosystem and its role in the world. In this paper, a cost-effective coral fish detection and identification method is proposed. Using the up-to-date models of Convolutional Neural Network (CNN), this paper is able to analyze an underwater footage and highlight all the detectable fish in color frames, and then identifying the species name among the detectable fish. Fish object detection was employed using Open Images Dataset and Tensorflow Object Detection. The paper further explores CNN with squeeze and excitation for fish classification. The proposed model was evaluated on fish4-knowledge dataset and achieved 100% validation accuracy after 50 epochs, better than AlexNet and RetNet, indicating that the solution is robust and practical. In addition, a dataset for coral fish classification in specific location was built using different sources. The model achieved 100% validation accuracy on the proposed dataset.

Comparative analysis of the morphology, karyotypes and biochemical composition of muscle in Siniperca chuatsi, Siniperca scherzeri and the F1 hybrid (S. chuatsi ♀ × S. scherzeri ♂)

Combining morphological comparison with karyotyping can make a more accurate and comprehensive study of mandarin fish (Siniperca spp.) and hybrid progenies.In recent years, the domestication of hybrid mandarin fish has become a hot spot in China, and it is essential to evaluate the nutritional value of the fish after food conversion. In this study, the F1 hybrid (S. chuatsi ♀ × S. scherzeri♂) were fed with artificial feed for five months. Afterwards, we studied the morphology, karyotypes, and biochemical composition of muscle in the F1 hybrid and its parents. The morphological results showed that total length and body height of the F1 hybrid appear to derive more from S. chuatsi than S. scherzeri. The DNA ploidy and karyotype results indicated that the F1 hybrid have a diploid chromosome number (2n = 48) and the same karyotypes (2n = 4sm + 14th+30t, NF = 52) with its parents, suggesting that the chromosome of the mandarin fish has convergence in the process of evolution. The results of biochemical composition of muscle showed that there were no significant difference in other nutritional indicators (P > 0.05), except the Gly (glycine) content of the F1 hybrid was significantly higher than that of S. chuatsi (P < 0.05) under the experimental feeding regime, suggesting that the nutritional value of the F1 hybrid with artificial feeding culture is also higher. The study is expected to provide a reference for further research on the domestication, classification, chromosomes, and nutritional requirements of mandarin fish.

Keeping it classy: classification of live fish and ghost PIT tags detected with a mobile PIT tag interrogation system using an innovative analytical approach

The ability of passive integrated transponder (PIT) tag data to improve demographic parameter estimates has led to the rapid advancement of PIT tag systems. However, ghost tags create uncertainty about detected tag status (i.e., live fish or ghost tag) when using mobile interrogation systems. We developed a method to differentiate between live fish and ghost tags using a random forest classification model with a novel data input structure based on known fate PIT tag detections in the San Juan River (New Mexico, Colorado, and Utah, USA). We used our model to classify detected tags with an overall error rate of 6.8% (1.6% ghost tags error rate and 21.8% live fish error rate). The important variables for classification were related to distance moved and response to monsoonal flood flows; however, habitat variables did not appear to influence model accuracy. Our results and approach allow the use of mobile detection data with confidence and allow for greater accuracy in movement, distribution, and habitat use studies, potentially helping identify influential management actions that would improve our ability to conserve and recover endangered fish.

Systematics and Distribution of Freshwater Icthyofauna in Babylon Province

Several studies have been carried out on the classification of freshwater fish (Icthyofauna) in all of Iraq or in the Tigris and Euphrates basin. The current study consisted of (82 specimens) were collected from three different regions within Babylon Province for seven months. Taxonomic status of specimens were studied depending on morphometric traits (biometrics), then preserved in 70% ethyl alcohol. Surveyed area included three main sites in Babylon province, which is located in the center of Iraq. The diagnostic results showed that the collected fish samples related to five families, nine genera, and twelve species, as following: Family cyprinidae (Aspius vorax, Barbus barbulus, B. luteus, B. xanthopterus, B. grypus, Carassius auratus, Cyprinus carpio); family Bagridae (Mystus pelusius); family Siluridae (Silurus triostegus); family Mugilidae (Planiliza) (Liza abu) and (L. subviridis); family Cichlidae (Tilapia zilli).

Systematic reappraisal of the anti-equatorial fish genus Microcanthus Swainson (Teleostei: Microcanthidae), with redescription and resurrection of Microcanthus joyceae Whitley.

The taxonomy and classification of the microcanthid fish genus Microcanthus Swainson has been a subject of contention dating back to the 19th century. Its allopatric, disjunct anti-equatorial distribution across the Indo-West Pacific has resulted in the recognition of several nominal taxa, though these have been widely regarded as synonyms of Microcanthus strigatus (Cuvier). Following the results published in a companion study elsewhere by the authors, the taxonomy of Microcanthus and the validity of these nominal synonyms are herewith revised. Microcanthus strigatus is redescribed on the basis of 66 specimens from East Asia, Hawaii and Western Australia, and M. joyceae is resurrected and redescribed on the basis of 25 specimens from eastern Australia and the southwest Pacific. Microcanthus differs from other microcanthid genera in having the following combination of characters: dorsal-fin rays XI,15-17 (usually XI,16); anal-fin rays III,13-15 (usually III,14); pectoral-fin rays 15-17 (usually 16); scales ctenoid with ctenial bases present; lateral-line scales partially or heavily obscured by adjacent scales; and body pale in preservation with five horizontal dark stripes reaching the posterior edges of dorsal and anal fins, and base of caudal fin. The review is accompanied by a key to the genera of Microcanthidae.