• Title/Summary/Keyword: 조류 탐지

Search Result 28, Processing Time 0.027 seconds

Implementation of Vision-based Wild Bird Detection and Repelling System using RaspberryPi (라즈베리파이를 활용한 비전기반 야생조류 침입 탐지 및 퇴치 시스템의 구현)

  • Lee, Cheol-won;Na, Daeyoung;Muminov, Azamjon;Karimov, Botirjon;Jeon, Heung Seok
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 2018.07a
    • /
    • pp.507-508
    • /
    • 2018
  • 본 논문에서는 라즈베리파이를 활용하여 야생조류의 행동에 반응하는 비전기반 야생조류 퇴치 장치를 구현하였다. 저가형 라즈베리파이 보드를 기반으로 카메라센서와 OpenCV 모션탐지기법을 활용하여 야생조류의 침입을 탐지하고, 그리고 야생조류가 소리별로 반응하는 데이터를 누적시키는 방법을 활용하여 효율적으로 퇴치하도록 개발하였다. 성능평가는 야생 직박구리와 박새를 포획하여 야외 실험장에서 진행하였고 실제 환경에서도 야생조류를 퇴치할 수 있다는 결과를 보여준다.

  • PDF

해상풍력발전기 조류환경 영향평가를 위한 인공지능 조류충돌방지 시스템

  • 이희용
    • Proceedings of the Korean Institute of Navigation and Port Research Conference
    • /
    • 2022.11a
    • /
    • pp.380-382
    • /
    • 2022
  • 해상풍력발전단지 환경평가를 위한 조류충돌저감장치를 개발하기 위하여, 천연기념물 조류를 구부할 수 있는 인공지능 카메라를 개발한다. 보호해야 할 조류를 90프로 이상 정확하게 구분하기 위한 계층구조 라벨링 방법을 고안하고 YOLO5 모델을 사용하여 학습을 수행하고, 그 결과를 보인다.

  • PDF

A Comparison of Bioacoustic Recording and Field Survey as Bird Survey Methods - In Dongbaek-dongsan and 1100-altitude Wetland of Jeju Island - (조류 조사 방법으로써 생물음향 녹음과 현장 조사의 비교 - 제주 동백동산과 1100고지 습지를 대상으로 -)

  • Se-Jun Choi;Kyong-Seok Ki
    • Korean Journal of Environment and Ecology
    • /
    • v.37 no.5
    • /
    • pp.327-336
    • /
    • 2023
  • This study aimed to propose an effective method for surveying wild birds by comparing the results of bioacoustic detection with those obtained through a field survey. The study sites were located at Dongbaek-dongsan and a 1100-altitude wetland in Jeju-do, South Korea. The bioacoustic detection was conducted over the course of 12 months in 2020. For the bioacoustic detection, a Song-meter SM4 device was installed at each study site, recording bird songs in 1-min per hour, .wav, and 44,100 Hz format. The findings of the field survey were taken from the 「Long-term trends of Bird Community at Dongbaekdongsan and 1100-Highland Wetland of Jeju Island, South Korea.」 by Banjade et al. (2019). The results of this study are as follows. First, the avifauna identified using bioacoustic detection comprised 29 families and 46 species in Dongbaek-dongsan, and 16 families and 25 species in the 1100-altitude wetland. Second, based on the song frequency, the dominant species in Dongbaek-dongsan were Hypsipetes amaurotis (Brown-eared Bulbul, 33.62%), Horornis diphone (Japanese Bush Warbler, 12.13%), and Zosterops japonicus (Warbling White-eye, 9.77%). In the 1100-altitude wetland the dominant species were Corvus macrorhynchos (Large-billed Crow, 27.34%), H. diphone (19.43%), and H. amaurotis (16.56%). Third, in the field survey conducted at Dongbaek-dongsan, the number of detected bird species was 39 in 2009, 51 in 2012, 35 in 2015, and 45 in 2018, while the bioacoustic detection identified 46 species. In the field survey conducted in the 1100-altitude wetland, the number of detected bird species was 37 in 2009, 42 in 2012, 34 in 2015, and 38 in 2018, while the bioacoustics detection identified 25 species. Overall, 43.6% of the 78 species detected in the field survey in Dongbaek-dongsan (34 species) were identified using bioacoustic detection, and 38.3% of the 47 species detected in the field survey in the 1100-altitude wetland (18 species) were identified using bioacoustic detection. Fourth, the bioacoustic detection identified 9 families and 12 species of birds in Dongbaek-dongsan, and 3 families and 7 species of birds in the 1100-altitude wetland. No results from field survey were available for these species. The identified birds were predominantly nocturnal, including Otus sunia (Oriental Scops Owl) and Ninox japonica (Northern Boobook), passage migrants, including Larvivora cyane (Siberian Blue Robin), L. sibilans (Rufous-tailed Robin), and winter visitors with a relatively small number of visiting individuals, such as Bombycilla garrulus (Bohemian Waxwing) and Loxia curvirostra (Red Crossbill). Fifth, the birds detected in the field survey but not through bioacoustic detection included 18 families and 48 species in Dongbaek-dongsan and 14 families and 27 species in the 1100-altitude wetland; the most representative families were Ardeidae, Accipitridae, and Muscicapidae. This study is significant as it provides essential data supporting the possibility of an effective survey combining bioacoustic detection with field studies, given the increasing use of bioacoustic devices in ornithological studies in South Korea.

Analysis of performance changes based on the characteristics of input image data in the deep learning-based algal detection model (딥러닝 기반 조류 탐지 모형의 입력 이미지 자료 특성에 따른 성능 변화 분석)

  • Juneoh Kim;Jiwon Baek;Jongrack Kim;Jungsu Park
    • Journal of Wetlands Research
    • /
    • v.25 no.4
    • /
    • pp.267-273
    • /
    • 2023
  • Algae are an important component of the ecosystem. However, the excessive growth of cyanobacteria has various harmful effects on river environments, and diatoms affect the management of water supply processes. Algal monitoring is essential for sustainable and efficient algae management. In this study, an object detection model was developed that detects and classifies images of four types of harmful cyanobacteria used for the criteria of the algae alert system, and one diatom, Synedra sp.. You Only Look Once(YOLO) v8, the latest version of the YOLO model, was used for the development of the model. The mean average precision (mAP) of the base model was analyzed as 64.4. Five models were created to increase the diversity of the input images used for model training by performing rotation, magnification, and reduction of original images. Changes in model performance were compared according to the composition of the input images. As a result of the analysis, the model that applied rotation, magnification, and reduction showed the best performance with mAP 86.5. The mAP of the model that only used image rotation, combined rotation and magnification, and combined image rotation and reduction were analyzed as 85.3, 82.3, and 83.8, respectively.

YOLO based Drone detection on Embeded Board (임베디드 보드에서의 YOLO 기반 드론 탐지)

  • Yu, ByeungHo;Park, HanBin;Kim, MinSung;Choi, Haechul
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • fall
    • /
    • pp.335-337
    • /
    • 2021
  • 최근 드론의 용도는 취미, 공연, 농업, 안전, 군사, 연구, 물자수송 등 다양한 분야와 목적으로 활용되고 있다. 더불어 드론의 불법적 활용으로 인한 안전 및 법적 문제 또한 빈번히 발생하고 있어, 이런 문제들을 예방하기 위한 드론의 탐지 기술이 활발히 연구되고 있다. 본 논문은 카메라로 촬영된 영상에서 조류와 같은 다른 객체와 구별하여 드론을 탐지하는 기술과 상공에서 바라본 객체들을 탐지하는 기술을 구현한다. 제안 방법은 딥러닝 기반의 YOLOv4를 사용하였다. UAV_123 데이터세트로 학습한 실험 결과, mAP는 85%, Recall은 85%, Precision은 81%의 정확도를 보였다. 제안 방법은 인명 구조, 배송, 건축 뿐만 아니라 안티 드론 시장에서도 효과적으로 활용될 수 있을 것으로 기대된다.

  • PDF

Artificial Intelligence-Based Harmful Birds Detection Control System (인공지능 기반 유해조류 탐지 관제 시스템)

  • Sim, Hyun
    • The Journal of the Korea institute of electronic communication sciences
    • /
    • v.16 no.1
    • /
    • pp.175-182
    • /
    • 2021
  • The purpose of this paper is to develop a machine learning-based marine drone to prevent the farming from harmful birds such as ducks. Existing drones have been developed as marine drones to solve the problem of being lost if they collide with birds in the air or are in the sea. We designed a CNN-based learning algorithm to judge harmful birds that appear on the sea by maritime drones operating by autonomous driving. It is designed to transmit video to the control PC by connecting the Raspberry Pi to the camera for location recognition and tracking of harmful birds. After creating a map linked with the location GPS coordinates in advance at the mobile-based control center, the GPS location value for the location of the harmful bird is received and provided, so that a marine drone is dispatched to combat the harmful bird. A bird fighting drone system was designed and implemented.

Deep Learning Based Floating Macroalgae Classification Using Gaofen-1 WFV Images (Gaofen-1 WFV 영상을 이용한 딥러닝 기반 대형 부유조류 분류)

  • Kim, Euihyun;Kim, Keunyong;Kim, Soo Mee;Cui, Tingwei;Ryu, Joo-Hyung
    • Korean Journal of Remote Sensing
    • /
    • v.36 no.2_2
    • /
    • pp.293-307
    • /
    • 2020
  • Every year, the floating macroalgae, green and golden tide, are massively detected at the Yellow Sea and East China Sea. After influx of them to the aquaculture facility or beach, it occurs enormous economic losses to remove them. Currently, remote sensing is used effectively to detect the floating macroalgae flowed into the coast. But it has difficulties to detect the floating macroalgae exactly because of the wavelength overlapped with other targets in the ocean. Also, it is difficult to distinguish between green and golden tide because they have similar spectral characteristics. Therefore, we tried to distinguish between green and golden tide applying the Deep learning method to the satellite images. To determine the network, the optimal training conditions were searched to train the AlexNet. Also, Gaofen-1 WFV images were used as a dataset to train and validate the network. Under these conditions, the network was determined after training, and used to confirm the test data. As a result, the accuracy of test data is 88.89%, and it can be possible to distinguish between green and golden tide with precision of 66.67% and 100%, respectively. It is interpreted that the AlexNet can be pick up on the subtle differences between green and golden tide. Through this study, it is expected that the green and golden tide can be effectively classified from various objects in the ocean and distinguished each other.

Application of Standardized North American Marsh Bird Monitoring Protocols to Survey Inconspicuous Marsh Birds in Korea (은둔형 습지 조류의 효과적인 조사 방법 탐색을 위한 국외 프로토콜의 시범 적용)

  • Lee, Sang-Yeon;Sung, Ha-Cheol
    • Korean Journal of Ecology and Environment
    • /
    • v.52 no.2
    • /
    • pp.143-150
    • /
    • 2019
  • Although inconspicuous marsh birds are an indicator of marsh health, there is little understanding of their status and population trends due to their behavioral characteristics and lack of reliable survey methods in Korea. We applied the Standardized North American Marsh Bird Monitoring Protocols(SNAMBMP) already validated in North America for effective survey of the marsh birds. We selected 29 sites with emergent marshes, rice fields and riparian forests in Seocheon-gun, Buyeo-gun and Gunsan-si. We conducted the survey with a combination of passive 5 minute point-count and vocal survey method (30 seconds call-broadcasting+30 seconds silence) that was targeted eight species 2~7 times/site from March to July 2017. Four species, Brown-cheeked Rail(Rallus indicus), Ruddy-breasted Crake (Porzana fusca), Watercock (Gallicrex cinerea) and Greater Painted-snipe (Rostatula benghalensis), were detected at one site respectively (naïve occupancy rate=0.035). Vocal survey method with conspecific call-broadcasting provided better on Brown-cheeked Rail and Watercock than the others. We suggest a combination of passive point-count and vocal survey method like SNAMBMP to monitor inconspicuous marsh birds at nationwide scale and collection of sound files through recording of the entire process during the survey.

Nocturnal Birds Detection and Ecological Characteristics through Bioacoustic Monitoring (생물음향 모니터링 기법을 이용한 야행성 조류 탐지 및 생태적 특성 분석)

  • Choi, Se-Jun;Ki, Kyong-Seok
    • Korean Journal of Environment and Ecology
    • /
    • v.33 no.6
    • /
    • pp.636-644
    • /
    • 2019
  • The purpose of this study was to investigate the callings of nocturnal birds using bioacoustic recording technology to identify species and to analyze the ecological characteristics of each species. Three sites - Seoraksan National Park, National Institute of Ecology, and Mudeungsan National Park - were investigated. The investigation period was from the middle of April 2018 to early March 2019 for Seoraksan national park, from late February of 2018 to the middle of February 2019 for the National Institute of Ecology, and from the middle of February 2018 to the end of August 2018 for Mudeungsan National Park. The main research results are as follows. Firstly, nocturnal bird species identified by the survey included Caprimulgus indicus, Otus sunia, Zoothera aurea, Bubo bubo, and Strix uralensis, 5 species in total. Secondly, the breeding call period of each species was from early May to early August for C. indicus, from early April to the end of September for O. sunia, from early March to early October for Z. aurea, from late September to early February for B. bubo, and from mid-January to early March for S. uralensis. Thirdly, the mating call rhythm was between 16:00 and 10:00 on the following day for all the observed species in the three regions, and the peak time zone was from 20:00 to 06:00 on the following day. Fourthly, there was no correlation between the cumulative call frequency and the precipitation for each species. Fifthly, the mean temperature during the period when the specific calls of nocturnal birds were detected was -4.00 ℃ for S. uralensis, 2.58 ℃ for B. bubo, 13.66 ℃ for Z. aurea, 19.50 ℃ for O. sunia, and 20.77 ℃ for C. indicus. The ANOVA results showed that there was a significant difference in mean temperature for the calling by species and that the mean temperature was S. uralensis, B. bubo, Z. aurea, and O. sunia-C. indicus, in the ascending order, for 4 groups in total. The period of the specific mating calls confirmed by the study is a period in which the frequency of calls was the highest among the periods when the specific calls were detected. Since it is associated with the known mating period of each species, the period of the high frequency of calls confirmed by the bioacoustic monitoring can be regarded as the mating season. This study is meaningful in that it is the early research that has used the bioacoustic recording technology to identify species and ecological characteristics of species of nocturnal birds in Korea.

Monitoring of Floating Green Algae Using Ocean Color Satellite Remote Sensing (해색위성 원격탐사를 이용한 부유성 녹조 모니터링)

  • Lee, Kwon-Ho;Lee, So-Hyun
    • Journal of the Korean Association of Geographic Information Studies
    • /
    • v.15 no.3
    • /
    • pp.137-147
    • /
    • 2012
  • Recently, floating green algae (FGA) in open oceans and coastal waters have been reported over wide area, yet accurate detection of these using traditional ground based measurement and chemical analysis in the laboratory has been difficult or even impossible due to the lack of spatial resolution, coverage, and revisit frequency. In contrast, spectral reflectance measurement makes it possible to quickly assess the chlorophyll content in green algae. Our objectives are to investigate the spectral reflectance of the FGA observed in the Yellow Sea and to develop a new index to detect FGA from satellite imagery, namely floating green algae index (FGAI), which uses relatively simple reflectance ratio technique. The Moderate Resolution Imaging Spectroradiometer (MODIS) and Geostationary Ocean Color Imager (GOCI) satellite images at 500m spatial resolution were utilized to produce FGAI which is defined as the ratio between reflectance at 860nm and 660nm bands. Both FGAI results yielded reasonable green algae detection at the regional scale distribution. Especially houly GOCI observations can present more detaield information of FGAI than low-orbit satellite.