Fish Images Research Articles

Real-time nondestructive monitoring of Common Carp Fish freshness using robust vision-based intelligent modeling approaches

In the current research, the potential of a novel method based on the artificial neural network was investigated to diagnose the freshness of common carp (Cyprinus carpio) during ice storage. Fish as an aquaculture product has high nutrients and low-fat content. So, people have consumed it as a safe and high-value foodstuff in their daily diet. Investigation of fish freshness is proposed as a significant issue in the aquaculture industry since fish spoils rapidly. The applied system of this study is comprised of the following steps: First, images of samples were captured and the pre-processing operation was done on the images. Then, particular channels including R, G, B, H, S, I, L*, a*, and b* were computed. Next, feature extraction was performed to obtain 6 types of texture features from each channel. Afterward, the hybrid Artificial Bee Colony-Artificial Neural Network (ABC-ANN) algorithm was applied to select the best features. Finally, the Support Vector Machine (SVM), K-Nearest Neighbor (K-NN) and Artificial Neural Network (ANN) algorithms as the most common methods were used to classify fish images. The best performance of the K-NN classifier was calculated in the k = 8 neighborhood size with the accuracy of 90.48. The best kernel function for the SVM algorithm was polynomial with C, sigma, and accuracy of 1, 2 and 91.52 percent, respectively. In this system, the input layer has consisted of 22 neurons based on the feature selection operation and 4 classes including most fresh, fresh, fairly fresh and spoiled have been used as the number of output layer. At the end, the best results of the MLP networks were achieved by LM learning algorithm and 6 neurons in the hidden layer with the 22–10–4 topology and accuracy of 93.01 percent. The achieved results demonstrate the high performance of the ANN classifier for evaluation of common carp freshness during ice storage as a rapid, accurate, non-destructive, real-time and automated method. It shows the potential of computer vision method in combination with artificial neural networks as an intelligent technique for evaluation of fish freshness.

Photafish system: An affordable device for fish photography in the wild.

This paper describes the Photafish System, an innovative technique for capturing high-quality digital images of live freshwater fishes and other aquatic organisms in the field. This system may improve and facilitate the visual documentation of such organisms to illustrate guidebooks, populate online databases and support scientific papers, particularly taxonomic ones. The system was designed and tested to produce Taxonomically Informative Photos (TIPs) of living specimens and employs an Aquarium Assembly for Photography (AAP) distinguished by its portability and low cost. Using Neotropical freshwater fishes as a model, we propose an optimized protocol for documenting the morphology of live specimens to support taxonomy, as well as awareness and conservation of freshwater biodiversity. This paper addresses the technical aspects and challenges of field studio photography in remote locations. Tricks gleaned from personal experiences with nature photography are also shared. Finally, we briefly discuss the advantages and weaknesses of our approach.

The Relation Between Young Children's False Statements and Response Latency, Executive Functioning, and Truth-Lie Understanding.

This study examined relations between children's false statements and response latency, executive functioning, and truth-lie understanding in order to understand what underlies children's emerging ability to make false statements. A total of 158 (2- to 5-year-old) children earned prizes for claiming that they were looking at birds even when presented with images of fish. Children were asked recall ("What do you have?"), recognition ("Do you have a bird/fish?"), and outcome ("Did you win/lose?") questions. Response latencies were greater when children were presented with fish pictures than bird pictures, particularly when they were asked recall questions, and were greater for false statements than for true statements, again when children were asked recall questions. Older but not younger children exhibited longer latencies when making false responses to outcome questions, which suggests that younger children were providing impulsive desire-based responses to the outcome questions. Executive functioning, as measured by the Stroop task, was not related to false statements. Children who were better at labeling statements as the truth or not the truth were more proficient at making false statements. The results support the proposition that the cognitive effort required for making false statements depends on the types of questions asked.

Some Observations on the Mosaic Pavements in Lod, Huqoq and Wadi Hamam

As a result of a detailed comparative analysis of individual figurative elements (images of fish) in antique mosaic pavements discovered around the settlements of Huqoq and Wadi Hamam in Lower Galilee,and taking the mode of their execution into account, we can state with certainty that the mosaic panels unearthed in this region are characterized by an unusual stylistic manner, which brings them closer to the pavements from Lod, but makes them starkly different from the ones found in any other area of the late Empire. The evident dissimilarity of these pavements allows us to consider them an isolated group and assume that they were all produced at a single highly creative mosaic center active in Galilee during the 3rd–5th centuries A.D. Such a conclusion seems logical, no matter which method of scientific analysis is employed. In the article we detail the technical and technological aspects of making the foundation under the mosaic pavement, and specify the sequence of steps in setting up the layers to secure the durability of the pavement.Of particular importance has been the issue of the underpainting on the nucleus, viewed as a counterpart of today’s cartoon, and also to the analysis of the technological process of how the painted ground hidden under the layer of tesserae over it revealed itself in the Lod mosaic pavement.

Method for segmentation of overlapping fish images in aquaculture

Individual fish segmentation is a prerequisite for feature extraction and object identification in any machine vision system. In this paper, a method for segmentation of overlapping fish images in aquaculture was proposed. First, the shape factor was used to determine whether an overlap exists in the picture. Then, the corner points were extracted using the curvature scale space algorithm, and the skeleton obtained by the improved Zhang-Suen thinning algorithm. Finally, intersecting points were obtained, and the overlapped region was segmented. The results show that the average error rate and average segmentation efficiency of this method was 10% and 90%, respectively. Compared with the traditional watershed method, the separation point is accurate, and the segmentation accuracy is high. Thus, the proposed method achieves better performance in segmentation accuracy and effectiveness. This method can be applied to multi-target segmentation and fish behavior analysis systems, and it can effectively improve recognition precision. Keywords: aquaculture, image processing, overlapping segmentation, corner detection, improved Zhang-Suen algorithm DOI: 10.25165/j.ijabe.20191206.3217 Citation: Zhou C, Lin K, Xu D M, Liu J T, Zhang S, Sun C H, et al. Method for segmentation of overlapping fish images in aquaculture. Int J Agric & Biol Eng, 2019; 12(6): 135–142.

Method for segmentation of overlapping fish images in aquaculture

Method for segmentation of overlapping fish images in aquaculture

Representation of freshwater aquaculture fish behavior in low dissolved oxygen condition based on 3D computer vision

Dissolved oxygen (DO) plays an important role in industrialized freshwater aquaculture. Such deficiencies such as the high cost of water-quality monitoring system and the failure to accurately monitor or describe aquaculture water-quality existed in freshwater aquaculture water-quality monitoring system. Here, a kind of representation method applied to characterize industrialized aquaculture fish behavior in different degrees of DO deficiency is based on three-dimensional (3D) Computer Vision. 3D coordinate values of aquaculture fishes in water acquired from 3D Computer Vision Device by processing aquaculture fish image are applied to represent such parameters as the average activity and height of aquaculture fish in water. This method for representing different behaviors of industrialized freshwater aquaculture fish under the condition of anoxia is realized by using these parameters and combing with the experience of aquaculture. The results show that the representation of industrialized freshwater aquaculture fish based on 3D Computer Vision System can be applied to describe industrialized aquaculture fish behavior and effectively compensate for the shortfall spatial location of aquaculture fish unable to acquire from 2D monitoring system, which is helpful for the accurate and reasonable control of DO in aquaculture.

QIM, CATA, and Word Association methods for quality assessment of flathead gray mullet (Mugil cephalus): Going beyond the trained panel

AbstractThis study aimed to verify how untrained assessors would perform the quality evaluation of the flathead gray mullet (Mugil cephalus) using Check‐All‐That‐Apply (CATA) (n = 44) and Word Association (WA) (n = 112) methods by comparing the results with those from trained assessors (n = 5–9), using the Quality Index Method (QIM). Descriptors from QIM were adapted and used as CATA descriptors for the evaluation of fish with 3, 8, 10, 15, and 17 days of storage in ice. Images of fish with 3, 6, 10, 15, and 19 days of ice storage were used as stimuli for WA. The quality indices obtained from QIM and CATA presented a high correlation with r = .94 (p = .018). Multiple Factor Analysis revealed a high similarity between CATA and QIM sample configurations. Appropriate descriptors were selected in CATA for fish stored for different periods. This resulted in a better discrimination of samples in comparison with WA. M. cephalus, with 15 days of storage in ice, was rated unacceptable for consumption by both trained and untrained assessors. These findings showed that even untrained assessors were able to evaluate fish freshness, and thus emphasize the importance of actions aimed at maintaining the quality of raw fish from capture to commercialization.Practical applicationsThe evaluation of fish quality by consumers has been considered tricky. Only more experienced fish consumers have confidence in their ability to judge the freshness of fish. However, our study showed that untrained individuals, confronted with the task of evaluating fish quality using both the CATA and the Word Association methods, performed very well. This allowed us to conclude that they were clearly able to evaluate freshness. Untrained individuals were neither stricter, nor more lenient than trained assessors in determining the shelf life of the flathead gray mullet. Our study joined many others that have concluded that CATA method—thanks to its easiness to perform—resulted in the discrimination of samples similar to that obtained with trained panels. These findings demonstrate that CATA and WA methods, designed to be applicable with untrained panels, were effective in the evaluation of fish freshness, thus providing a solution when trained assessors are not available. In addition, the importance of trading fish with high quality is clear, so that the consumers can receive value for their money and feel satisfied.

Updated catalogue of bony fishes observed in deep waters at Isla del Coco National Park and Las Gemelas Seamount, Costa Rica (Eastern Tropical Pacific)

From 2006 to date the submersible DeepSee has been used to study the deep waters in and around Isla del Coco National Park, Costa Rica. Over these years, images and samples have been collected at depths between 50 and 450 m. Here we present a catalogue of bony fishes recorded by the submersible in deep waters of Isla del Coco, 500 km south-southwest of mainland Costa Rica, and at Las Gemelas Seamount within the designated Seamounts Management Area, 50 km southwest of Isla del Coco. A database with video-images of bony fishes was created from videos taken by the submersible’s high-definition digital camera from 2006 to 2015. Additional information on the distribution of fishes was obtained from 11 dives (24.3 hrs) using the remotely operated vehicle Hela at Las Gemelas Seamount during February 2012. Images of bony fishes were obtained during 376 dives (365 DeepSee dives plus 11 Hela dives) in 18 different locations, and here we report on a total of 85 taxa (i.e. putative species). In this catalogue we present images that are the first color photographs published for some species. In other cases, identifications were possible only to the level of genus (11), family (5) or order (1); hence the information is presented in terms of putative taxa. Four new records are reported for Isla del Coco: Leptenchelys vermiformis (Ophichthidae), Hyporthodus mystacinus (Serranidae), Kathetostoma averruncus (Uranoscopidae), and Symphurus diabolicus (Cynoglossidae). Depth ranges of twenty-six species are expanded beyond previously published records; three were observed in shallower water, twenty-two in deeper water and one was observed both shallower and deeper than in previous reports. As might be expected, increased bottom time with the submersible resulted in additions to the list of documented species. This highlights the importance of maintaining systematic research efforts in the deep waters around Isla del Coco, both for scientific purposes as well as conservation.

Differences between stance and foot preference evident in Osprey (Pandion haliaetus) fish holding during movement.

BackgroundSkateboarders, snowboarders, and surfers all show stance preferences for which foot is forward while moving. We are unaware of other animals than humans with a stance preference, perhaps excepting Osprey, who fly their caught fish beneath them in a foot‐forward stance. We hypothesize there should be no difference between left foot forward, right foot back (conventional) versus right foot forward left foot back (goofy) stances or for fish holding with unilateral left or right foot. Online, publicly available, convenience images of Osprey catching fish were accessed and assessed by five independent reviewers using different Internet search engines or online photo series. Stance preference and footedness were tested using chi‐square analysis.ResultsStance preferences were evident with the left foot forward (conventional stance) on average 64–78% of the time (all p < 0.02). No difference in foot preference for either one‐foot grabs of fish during flight or for non‐flight nest/perch fish holding was evident.ConclusionFlight stance of Osprey holding fish shows a lateralized preference in a proportion similar to skateboarders of surfers. We discuss stance preferences in the setting of complex movements and potential flight and survival advantages for Osprey.

A Deep learning method for accurate and fast identification of coral reef fishes in underwater images

Identifying and counting fish individuals on photos and videos is a crucial task to cost-effectively monitor marine biodiversity, yet it remains difficult and time-consuming. In this paper, we present a method to assist the identification of fish species on underwater images, and we compare our model performances to human ability in terms of speed and accuracy. We first tested the performance of a convolutional neural network (CNN) trained with different photographic databases while accounting for different post-processing decision rules to identify 20 fish species. Finally, we compared the performance of species identification of our best CNN model with that of humans on a test database of 1197 fish images representing nine species. The best CNN was the one trained with 900,000 images including (i) whole fish bodies, (ii) partial fish bodies and (iii) the environment (e.g. reef bottom or water). The rate of correct identification was 94.9%, greater than the rate of correct identification by humans (89.3%). The CNN was also able to identify fish individuals partially hidden behind corals or behind other fish and was more effective than humans to identify fish on smallest or blurry images while humans were better to identify fish individuals in unusual positions (e.g. twisted body). On average, each identification by our best CNN using a common hardware took 0.06 s. Deep Learning methods can thus perform efficient fish identification on underwater images and offer promises to build-up new video-based protocols for monitoring fish biodiversity cheaply and effectively.

What kind of effects do Fe2O3 and Al2O3 nanoparticles have on anaerobic digestion, inhibition or enhancement?

Fe2O3 and Al2O3 nanoparticles are widely used in products and find their way to wastewater treatment plants through the contact of water with these products. In this study, impacts of Fe2O3 and Al2O3 nanoparticles on methane potential of waste activated sludge (WAS) were investigated by comparing long and short term toxicity test results, modelling and FISH analysis. Methane production from the samples treated with the maximum concentration of Fe2O3 nanoparticles decreased 28.9% at the end of the long term BMP test. EC50 value for BMP test of the Fe2O3 nanoparticles was calculated as 901.94 mg/gTS with high coefficient of determination. Methane production from the samples treated with Al2O3 nanoparticles increased up to 14.8% (p > 0.05) at the end of the BMP test. However, short term toxicity tests for Fe2O3 and Al2O3 nanoparticles showed no impact on anaerobic digestion of WAS. Kinetic parameters obtained from models and captured FISH images were consistent with these results. Different impacts of nanoparticles on methane production suggested that anaerobic microorganisms can be affected from nanoparticles in various mechanisms. Hydrolysis (kH) and overall reaction rates (kR) values were determined as 0.0277 and 0.1441 d−1, respectively for each concentration of Al2O3 nanoparticles and raw WAS. Similarly, methane production from WAS containing 5, 50, 150 and 250 mgFe2O3/gTS were modeled with same kinetic values. However, kH constant was calculated as 0.0149 d−1 for 500 mgFe2O3/gTS. This means that Fe2O3 nanoparticles starting from this concentration inhibited the methanogenic consortium and caused decreased biogas production and spesific methane production rate.

ELECTRONIC ATLAS OF TYPICAL ACOUSTIC IMAGES FOR COMMERCIAL FISH SPECIES IN THE FAR-EASTERN SEAS OF RUSSIA

The first version of the software is developed for presentation of metaarchive and library of typical acoustic images of marine organisms for their size and species identification on echograms. It displays certain typical acoustic images of commercial fishes in the Far- Eastern Seas and their descriptions as electronic atlas and allows to choose a typical acoustic image by area (using its coordinates and/or name of fishery district), time of day, season, and object (species and size). For this atlas, the echograms of different fish species from the Bering, Chukchi, Okhotsk Seas and adjacent Pacific waters accompanied with the data on location, time, and results of trawling were accumulated since 1996. The options «Filter», «Echogram parameters», «Trawling parameters», «List of echograms», «Catch», «Size composition», «Maturity», and «Echogram image» are presented on the window of this atlas. The «Filter» limits the area of echograms search by coordinates, borders of fishery district, time of day, season, species and size of fish. The «Echogram parameters» allows to select the echogram palette suitable for the type of echosounder and gives information on time and distance of the record. The atlas could be applied in trawl-acoustic surveys and for maintenance of fisheries in the Far-Eastern Seas, as well as a guide for work with scientific and fishery sonar devices.

MOTIVES OF FISH AND DOLPHIN IN THE EASTERN EUROPEAN SCYTHIAN ANIMAL STYLE: TYPOLOGY, CHRONOLOGY, ORIGINS

As the statistical analysis shows, the images of Fish and Dolphin constitute an extremely insignificant share of the general array of images of the Eastern European Scythian animal style. According to the author’s data for 2017, the total array of the original images of this local variant of the Scythian-Siberian animal style consisted of 2207 imageries, without taking into account copies and «mirror» images. Meanwhile there are only 27 original images of Fish and 4 original images of Dolphin, correspondingly only 1.23 and 0.18 % of this array. However, despite all the marginality of the Fish theme and especially of the Dolphin one, the presence of these motifs related to the water element is quite noticeable in the Scythian art. These are both real images (Fish, Dolphin) and syncretic images integrating fish motif (Horned fish, Hippocampus-horse, Griffin-hippocampus, «Cock-hippocampus»). Almost all of these images belong to the era of «Scythian classics» (V—IV centuries BC), especially to the IV century BC — that is, to the time of the most intense influence of Greek art on the Scythian animal style. All this corresponds to the conclusions of S. S. Bessonova and D. S. Raevsky concerning the important role of water element in Scythian religious and mythological ideas.

Developing an Ornamental Fish Warehousing System Based on Big Video Data

We have developed an ornamental fish warehousing (OFWare) system based on big video data. The system is an application paradigm of information and communication technologies for traditional industries, specifically in the fields of aquaculture and agriculture. Live creatures are the main products of these industries, raising challenges for warehouse management. Warehousing of high unit-price ornamental fishes such as koi, stingray, and arowana is even more difficult since, in addition to counting and classification, such warehousing requires the identification of individual animals whose shapes and texture patterns vary as they grow. Therefore, rather than using invasive RFID-based systems, we combine mobile cloud computing and big data analytics techniques including image and video collection and transmission using handheld mobile devices, unsupervised texture pattern classification of fish tank videos, fish image retrieval, and statistical analysis. The proposed system is scalable based on a Hadoop framework and a small set of a single name-nodes and data-nodes can identify a particular fish among 500,000 koi in 7 seconds. The proposed warehousing system can form the basis for the development of breeding histories, anti-forgery certificate, and aquaculture business intelligence.

A biometric-based model for fish species classification

Fish identification is crucial for the survival of our threatened fish species. In this paper, a novel and robust biometric-based approach was proposed to identify fish species. The proposed approach consists of three phases. In the first phase, different features were extracted from fish images. In this phase, Weber's Local Descriptor (WLD) and color moments were used to extract texture and color features, respectively. Due to the high dimensionality of WLD features, in the second phase, Linear Discriminant Analysis (LDA) was applied to reduce the number of features and to discriminate between different classes. In the third phase, the AdaBoost classifier was used to identify fish species. We have collected a dataset that consists of four classes/species. To validate the results of the AdaBoost classifier, a comparison between three well-known classifiers (Naive Bayesian, k-Nearest Neighbor, and Multilayer Perceptron) was performed. The experimental results proved that our approach achieved excellent results (approximately 96.4%). Moreover, our model has been tested against different real challenges such as image rotation and image translation, and the proposed model achieved promising results.

Pengukuran Badan Ikan Berupa Estimasi Panjang, Lebar, dan Tinggi Berdasarkan Visual Capture

Currently, fish measurement process is done manually using the gauge which can cause inappropriate, ineffective result, and requires long time to finish, especially for a great amount of fish. Therefore, an automatic fish body measurement technique in the form of estimation of length, width, and height of fish based on visual capture is needed to facilitate fish body measurement to become more effective and efficient. This research uses five samples of fish in measuring length, width, and height manually to obtain the average data/value as the calibration of reference value for calculation process in the system and to be stored in the database. The stages begin with capturing fish image using a digital camera. Then, preprocessing stage was carried out to get the grayscale image of the fish. The object of the grayscale image was then segmented to separate the important and unimportant part of the object. Lastly, feature extraction process of the fish body from the calibration average value was carried out, and the estimated value of length, width, and height of fish are obtained automatically. The results show that the automatic measurement technique of fish body based on visual capture was able to produce the truth degree of accuracy of 80% to 95%.

AUTOMATIC DETERMINATION OF SEEDS FOR RANDOM WALKER BY SEEDED WATERSHED TRANSFORM FOR TUNA IMAGE SEGMENTATION

Tuna fish image classification is an important part to sort out the type and quality of the tuna based upon the shape. The image of tuna should have good segmentation results before entering the classification stage. It has uneven lighting and complex texture resulting in inappropriate segmentation. This research proposed method of automatic determination seeded random walker in the watershed region for tuna image segmentation. Random walker is a noise-resistant segmentation method that requires two types of seeds defined by the user, the seed pixels for background and seed pixels for the object. We evaluated the proposed method on 30 images of tuna using relative foreground area error (RAE), misclassification error (ME), and modified Hausdroff distances (MHD) evaluation methods with values of 4.38%, 1.34% and 1.11%, respectively. This suggests that the seeded random walker method is more effective than exiting methods for tuna image segmentation.

Improving Transfer Learning and Squeeze- and-Excitation Networks for Small-Scale Fine-Grained Fish Image Classification

Scientific studies on species composition and abundance distribution of fishes have considerable importance to the fishery industry, biodiversity protection, and marine ecosystem. In these studies, fish images are typically collected with the help of scuba divers or autonomous underwater vehicles. These images are then annotated manually by marine biologists. Such a process is certainly a tremendous waste of manpower and material resources. In recent years, the introduction of deep learning has helped making remarkable progress in this area. However, fish image classification can be considered as fine-grained problem, which is more challenging than common image classification, especially with low-quality and small-scale data. Meanwhile, well-known effective convolutional neural networks (CNNs) consistently require a large quantity of high-quality data. This paper presents a new method by improving transfer learning and squeeze-and-excitation networks for fine-grained fish image classification on low-quality and small-scale datasets. Our method enhances data augmentation through super-resolution reconstruction to enlarge the dataset with high-quality images, pre-pretrains, and pretrains to learn common and domain knowledge simultaneously while fine-tuning with professional skill. In addition, refined squeeze-and-excitation blocks are designed to improve bilinear CNNs for a fine-grained classification. Unlike well-known CNNs for image classification, our method can classify images with insufficient low-quality training data. Moreover, we compare the performance of our method with commonly used CNNs on small-scale fine-grained datasets, namely, Croatian and QUT fish datasets. The experimental results show that our method outperforms popular CNNs with higher fish classification accuracy, which indicates its potential applications in combination with other newly updated CNNs.

The Relation between Young Children's False Statements and Response Latency, Executive Functioning, and Truth-Lie Understanding

This study examined relations between children’s false statements and response latency, executive functioning, and truth-lie understanding in order to understand what underlies children’s emerging ability to make false statements. A total of 158 (2- to 5-year-old) children earned prizes for claiming that they were looking at birds even when presented with images of fish. Children were asked recall (“what do you have?”), recognition (“do you have a bird/fish?”), and outcome (“did you win/lose?”) questions. Response latencies were greater when children were presented with fish pictures than bird pictures, particularly when they were asked recall questions, and were greater for false statements than for true statements, again when children were asked recall questions. Older but not younger children exhibited longer latencies when making false responses to outcome questions, suggesting that younger children were providing impulsive desire-based responses to the outcome questions. Executive functioning, as measured by the day-night Stroop task, was not related to false statements. Children who were better at labeling statements on a truth-lie identification task were more proficient at making false statements. The results support the proposition that the cognitive effort required for making false statements is dependent upon the types of questions asked.