• Journal of information and communication convergence engineering
• /
• v.18 no.2
• /
• pp.106-114
• /
• 2020

We propose appropriate criteria for obtaining fish species data and number of learning data, as well as for selecting the most appropriate convolutional neural network (CNN) to efficiently classify exotic invasive fish species for their extermination. The acquisition of large amounts of fish species data for CNN learning is subject to several constraints. To solve these problems, we acquired a large number of fish images for various fish species in a laboratory environment, rather than a natural environment. We then converted the obtained fish images into fish images acquired in different natural environments through simple image synthesis to obtain the image data of the fish species. We used the images of largemouth bass and bluegill captured at a pond as test data to confirm the effectiveness of the proposed method. In addition, to classify the exotic invasive fish species accurately, we evaluated the trained CNNs in terms of classification performance, processing time, and the number of data; consequently, we proposed a method to select the most effective CNN.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.1
• /
• pp.39-46
• /
• 2019

In this study, before system development for the elimination of foreign fish species, we propose an algorithm to classify fish species by training fish images with CNN. The raw data for CNN learning were directly captured images for each species, Dataset 1 increases the number of images to improve the classification of fish species and Dataset 2 realizes images close to natural environment are constructed and used as training and test data. The classification performance of four CNNs are over 99.97% for dataset 1 and 99.5% for dataset 2, in particular, we confirm that the learned CNN using Data Set 2 has satisfactory performance for fish images similar to the natural environment. And among four CNNs, AlexNet achieves satisfactory performance, and this has also the shortest execution time and training time, we confirm that it is the most suitable structure to develop the system for the elimination of foreign fish species.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.48 no.2
• /
• pp.207-220
• /
• 2015

This study identified the species-specific, frequency-dependent characteristics of broadband acoustic scattering that facilitate classifying fish species using the pulse compression (PC) technique. Controlled acoustic scattering laboratory experiments were conducted with nine commercially important fish species using linear chirp signals (95-220 kHz) over an orientation angle range of ${\pm}45^{\circ}$ in the dorsal plane at approximately $1^{\circ}$ increments. The results suggest that the angular-dependent characteristics of the broadband echoes and the frequency-dependent variability in target strength (TS) were useful for inferring the fish species of interest. The scattering patterns in the compressed pulse output were extremely complex due to morphological differences among fish species, but the x-ray images strongly suggested that spatial separation correlated well with scattering for the head, skeleton, bone, otoliths, and swim bladder within each specimen.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.464-465
• /
• 2018

Bass and Bluegill, which are representative ecosystem disturbance species, are reported to be the most important factor in the reduction of domestic native fish populations in Korea. Therefore, it is necessary to develop system and field application technology for the extermination of these foreign species. Recently, the CNN(Convolutional Neural Network), one of the deep learning systems for the recognition, classification, and learning, has shown excellent performance. However, CNN data used for object recognition and classification were mainly applied to recognition and classification of other objects with distinct characteristics. This study proposes a system that applies CNN to the classification of fish species with similar characteristics.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.48 no.2
• /
• pp.221-232
• /
• 2015

Broadband echoes measured in live chub mackerel Scomber japonicus, goldeye rockfish Sebastes thompsoni, and fat greenling Hexagrammos otakii with different morphologies and internal characteristics were analyzed in time and frequency domains to understand the species-specific echo feature characteristics for classifying fish species. The mean echo image for each time-frequency representation dataset obtained as a function of orientation angle was extracted to mitigate the effect of fish orientation on acoustic scattering. The joint time-frequency content of the broadband echo signals was obtained using the smoothed pseudo-Wigner-Ville distribution (SPWVD). The SPWVDs were analyzed for each echo signature of the three fish species. The results show that the time-frequency analysis provided species-specific echo structure patterns and metrics of the broadband acoustic signals to facilitate fish species classification.

• Korean Journal of Ecology and Environment
• /
• v.51 no.2
• /
• pp.149-159
• /
• 2018

Twenty-nine wetlands (20 river type and 9 lake type wetlands) in Gyeongsangnam-do were investigated to understand the characteristics of fish assemblages by the wetland type and to suggest management strategies. As a result, $10.3{\pm}4.8$ species were collected from river type wetlands on average (${\pm}SD$) and $9.1{\pm}4.1$ species from lake type wetlands. Thus, there was no significant difference in the number of species between them (Mann-Whitney U test, P>0.05). However, the species that constitute the fish assemblage showed statistically significant differences between the two wetland types (PERMANOVA, Pseudo-F=2.9555, P=0.007). Furthermore, the species that contribute the most to each type of fish assemblage were Zacco koreanus (river type, 28.51%) and Lepomis macrochirus (lake type, 23.21%), respectively (SIMPER). The results of the NMDS analysis using the fish assemblage by place classified the species into three groups (river type, lake type, and others). The current wetland management is only focused on endangered species, but this study shows a difference in fish assemblage by wetland type. Therefore, a management system based information on endemic species, exotic species and major contribution species should be provided. Furthermore, the classification of some types of wetlands based on the present topography was found to be ambiguous, and wetland classification using living creatures can be used as a complementary method. This study has limitations because only two types of wetlands were analyzed. Therefore, a detailed management method that can represent every type of wetland should be prepared through the research of all types of wetlands in the future.

• The Journal of Fisheries Business Administration
• /
• v.53 no.4
• /
• pp.51-67
• /
• 2022

The Korean tariff rate for fishery products is a single tax rate of 10% for live fish and frozen seafood, and 20% for all others. Since FTAs have been concluded with several countries, the tariffs is not an appropriate means to protect domestic fishery producers. The differential tariff rate according to the scientific name (genus) of the fishery products, which was implemented 30 years ago to protect fishery products produced in the Korean coastal waters has lost its original purpose. It seems that future fishery trade policy should focus on IUU prevention, hygiene and safety of consumers rather than protecting fishery producers through customs tariffs. This paper suggest that a paradigm shift in the fishery producers protection policies such as direct financial support from the state, protection and development of fishery resources, and support for fostering the 6th industry rather than indirect protection through tariffs.

• The Journal of Fisheries Business Administration
• /
• v.43 no.1
• /
• pp.35-48
• /
• 2012

Inverse demand models are well established as market demands in theory and practice of the existing literature. However, the derivation and its interpretation of individual demands from the market demands are not well known in the literature. This paper analyzes the fish market in Korea by the inverse demand model and shows how we deduce the consumer's responses from the market responses when the markets determine the prices by the quantities demanded. It illustrates empirically how this can be done applying to the korean fish market data. The empirical results show that all fishes are price inflexible and mackerels and hairtails are scale flexible in the market demand while mackerels, hairtails, and croakers are price elastic and mackerels and hairtails are income inelastic in the individual demand. The methodology and empirics used in the paper will make a contribution to the existing literature especially for the purpose of recovering consumer's demand from the market demand, thus implementing the policies to administer the fish markets.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.30 no.2
• /
• pp.147-156
• /
• 2024

The sustainable management and enhancement of marine resources are becoming increasingly important issues worldwide. This study was conducted in response to these challenges, focusing on the development and performance comparison of fish detection and classification models as part of a deep learning-based technique for assessing the effectiveness of marine resource enhancement projects initiated by the Korea Fisheries Resources Agency. The aim was to select the optimal model by training various sizes of YOLOv8-Seg models on a fish image dataset and comparing each performance metric. The dataset used for model construction consisted of 36,749 images and label files of 12 different species of fish, with data diversity enhanced through the application of augmentation techniques during training. When training and validating five different YOLOv8-Seg models under identical conditions, the medium-sized YOLOv8m-Seg model showed high learning efficiency and excellent detection and classification performance, with the shortest training time of 13 h and 12 min, an of 0.933, and an inference speed of 9.6 ms. Considering the balance between each performance metric, this was deemed the most efficient model for meeting real-time processing requirements. The use of such real-time fish detection and classification models could enable effective surveys of marine resource enhancement projects, suggesting the need for ongoing performance improvements and further research.

• Journal of fish pathology
• /
• v.13 no.2
• /
• pp.75-86
• /
• 2000

Monogeneans(Phylum Platyhelminthes) have been known as common parasites onto the skin and gills of cultured freshwater fishes. Infestation with these parasites result in a great loss in aquaculture industry. Some classification studies on these parasites have mostly been conducted in Japan but rarely in Korea. For the purpose of classifying monogenean parasites infesting two Korean freshwater fishes, eel (Anguilla japonica) and catfish(Parasilurus asotus), samples captured from March 1998 to April 2000 were examined. Here we report for the first time in Korea that four different species of monogeneans are identified: Pseudodactylogyrus bini and P. anguillae in eels; Ancylodiscoides infundibulovagina and Ancylodiscoides sp. in catfish.