• 수산해양기술연구
• /
• 제50권4호
• /
• pp.520-529
• /
• 2014

Experiment was set up to analyze bycatch and discards situation including snow crap Chionoecetes opilio of whelk trap. Four types of trap were used: drum type trap with PE net; drum type trap with PBS net; cylinder type trap with PE net; and cylinder type net with PBS net. Three funnels were attached in drum type trap and two funnels were used in cylinder type trap. A fleet of traps was consisted with one hundred traps. 25 traps of each type were set on a line in repeated sequence. Field experiments were conducted with 6 fishing operations in the Uljin waters, East Sea in July 2014. Catch of target whelks were 173,261 g and catch rate was 48.7% of total catch, while bycatch were 182,571 g, 51.3% of tatal catch. The catch rate of bycatch was 2.6% higher than that of target catch. Bycatch weight of snow crap was the highest as 142,987 g and formed about 40.2% of total catch, followed giant octopus, Enteroctopus dofleini, 31,762 g (8.9%). Bycatch rate of cylinder type trap was 2.3 times higher than that of drum type trap. Discard rate (discard/(discard+landing)) was 43.6%. Discard rate was the highest at cylinder type trap with PBS net as 63.1%, followed cylinder type trap with PE net as 47.9%, drum type trap with PE net as 33.4%, the lowest at drum type trap with PBS net as 22.1%.

• 수산해양기술연구
• /
• 제32권1호
• /
• pp.7-15
• /
• 1996

Full scale experiment was carried out in the southern sea of Korea to compare some important factors tested in the model experiment. The results obtained can be summarized as follows ; 1. The changing aspect of mouth performance of the full scale net was almost coincided with the results obtained by the model experiment. The vertical opening(H) and the opening area(S) can be expressed as a function of the towing velocity(V) as H=48.78. $e^0.38V$(unit: m, k't) S= 1,443 .$e^-0.25V$(unit: $m^2V$, k't) 2. The changing aspect of working depth of the full scale net was almost coincided with the results obtained by the model experiment. The depth(D) can be expressed as a function of the towing velocity(V) and the warp length(L) as D=92.49.$V^1.37$(unit: m, k't, L= 150m) D= 12.07+0.32. L (unit: m, V=2k't) [)= - 7.90+0.22 . L (unit: m, V=3k't) 3. Some problems were found to operate A - type full scale net by common bottom pair trawlers. The problems can be summarized as follows; (1) Entangling of wing and square head ropes while net casting.(2) Man power needed and time spent for net hauling by common bottom trawlers increased considerably.( 3) Tearing of nettings caused by over -load of tension and entangling of net pendant while net hauling. To solve these problems, the trawlers are favorable to be equipped with variable pitch propeller and llet drum. While the net is favorable to be constructed with trifurcated net pendant in stead of quadrifurcated net pendant used at present.

• Journal of Ecology and Environment
• /
• 제38권3호
• /
• pp.383-388
• /
• 2015

Reservoirs consist of two different environments, the littoral and the pelagic zone, and different fishing gear is commonly used in each zone-gill nets in the pelagic zone and electrofishing in the littoral zone. However, an active fishing gear, the cast net, is normally used instead of electrofishing for scientific studies in South Korea. In order to estimate cast net effectiveness for determining fish status in reservoirs, the study was conducted at 15 reservoirs with two different fishing gears: a cast net in the littoral zone and gill nets in the pelagic zone. When combining catches of both gears, species richness increased substantially compared to using one gear only. There was a size difference in fish caught by each net, and small fish were predominantly caught with the cast net due to its small mesh size (7 mm). The combined length of six species, used for length-weight relationship analysis, collected with the cast net was smaller than that collected with gill nets (independent t-test, P < 0.05). In this study, cast net sampling provided sufficient data for the littoral zone, but not enough to identify the overall fish assemblage in studied reservoirs. Utilization of only one gear can therefore lead to substantial underestimation of fish status, and a combination of both gears is recommended for determining more reliable estimates of fish status in reservoirs.

• 대한임베디드공학회논문지
• /
• 제15권5호
• /
• pp.227-234
• /
• 2020

In this paper, we propose wav-U-Net to improve speech enhancement in heavy noisy environments, and it has implemented three principal techniques. First, as input data, we use 128 modified Mel-scale filter banks which can reduce computational burden instead of 512 frequency bins. Mel-scale aims to mimic the non-linear human ear perception of sound by being more discriminative at lower frequencies and less discriminative at higher frequencies. Therefore, Mel-scale is the suitable feature considering both performance and computing power because our proposed network focuses on speech signals. Second, we add a simple ResNet as pre-processing that helps our proposed network make estimated speech signals clear and suppress high-frequency noises. Finally, the proposed U-Net model shows significant performance regardless of the kinds of noise. Especially, despite using a single channel, we confirmed that it can well deal with non-stationary noises whose frequency properties are dynamically changed, and it is possible to estimate speech signals from noisy speech signals even in extremely noisy environments where noises are much lauder than speech (less than SNR 0dB). The performance on our proposed wav-U-Net was improved by about 200% on SDR and 460% on NSDR compared to the conventional Jansson's wav-U-Net. Also, it was confirmed that the processing time of out wav-U-Net with 128 modified Mel-scale filter banks was about 2.7 times faster than the common wav-U-Net with 512 frequency bins as input values.

• 인터넷정보학회논문지
• /
• 제18권5호
• /
• pp.61-67
• /
• 2017

본 논문에서는 딥러닝의 회선신경망을 이용한 실제 소셜 네트워크 상의 이미지 분류가 얼마나 효과적인지 알아보기 위한 실험을 수행하고, 그 결과와 그를 통해 알게 된 교훈에 대해 소개한다. 이를 위해 ImageNet Large Scale Visual Recognition Challenge(ILSVRC)의 2012년 대회와 2015년 대회에서 각각 우승을 차지한 AlexNet 모델과 ResNet 모델을 이용하였다. 평가를 위한 테스트 셋으로 인스타그램에서 수집한 이미지를 사용하였으며, 12개의 카테고리, 총 240개의 이미지로 구성되어 있다. 또한, Inception V3모델을 이용하여 fine-tuning을 실시하고, 그 결과를 비교하였다. AlexNet과 ResNet, Inception V3, fine-tuned Inception V3 이 네 가지 모델에 대한 Top-1 error rate들은 각각 49.58%, 40.42%, 30.42% 그리고 5.00%로 나타났으며, Top-5 error rate들은 각각 35.42%, 25.00%, 20.83% 그리고 0.00%로 나타났다.

• 한국농림기상학회지
• /
• 제5권2호
• /
• pp.128-137
• /
• 2003

순복사는 지표 에너지 수지의 가장 근본적인 요소 중 하나이다. 순복사의 정확한 관측을 위해, 주기적이고 지속적인 순복사계 보정이 요구된다. 플럭스 관측에 널리 사용되는, 두 가지 타입의 대표적인 순복사계 (Q-7.1과 CNR1)의 상호 비교 및 보정 실험이 약 4개월 간격으로 두 차례 시행되었다. Q-7.1과 CNR1 간의 차이는 7.7% 이내였고, 표준 기기와의 보정 후 오차는 3.2%이내였다. 순복사계의 반응 차이와 보정 계수는 대기 상태, 특히 계절 변화에 따른 온도 차이에 따라 다르게 나타났다. 결론적으로, 주기적으로 보정된 Q-7.1은 CNR1을 대체하여 장기 관측에 사용될 수 있고, 보정 주기로는 4-6개월이 권장된다.

• 한국음향학회지
• /
• 제39권5호
• /
• pp.454-460
• /
• 2020

음성인식 기능을 제공하는 인공지능 비서들은 정확도가 뛰어난 클라우드 기반의 음성인식을 통해 동작한다. 클라우드 기반의 음성인식에서 시작 단어 인식은 대기 중인 기기를 활성화하는 데 중요한 역할을 한다. 본 논문에서는 공개 데이터셋인 구글의 Speech Commands 데이터셋을 사용하여 스펙트로그램 및 멜-주파수 캡스트럼 계수 특징을 입력으로 하여 모바일 기기에 대응한 저 연산 시작 단어 검출을 위한 합성곱 신경망의 성능을 비교한다. 본 논문에서 사용한 합성곱 신경망은 다층 퍼셉트론, 일반적인 합성곱 신경망, VGG16, VGG19, ResNet50, ResNet101, ResNet152, MobileNet이며, MobileNet의 성능을 유지하면서 모델 크기를 1/25로 줄인 네트워크도 제안한다.

• 한국임상수의학회지
• /
• 제37권5호
• /
• pp.249-254
• /
• 2020

Neutrophil extracellular trap (NET) formation is an immune response for the invasion of microbes. The purpose of this study is to examine the effect of zinc on NET formation of porcine peripheral blood polymorphonuclear cells (PMNs). The NET formation of PMNs was measured by fluorescence microplate reader. The production of tumor necrosis factor (TNF)-α in the culture supernatants from zinc-treated peripheral blood mononuclear cells (PBMCs) was measured by enzyme-linked immunosorbent assay (ELISA). Zinc itself did not have no effect on NET formation. However, the NET formation of PMNs was increased by culture supernatants from PBMCs treated with zinc. Also, the NET formation of PMNs was increased by recombinant porcine (rp) TNF-α. The production of TNF-α in PBMCs culture supernatants was shown to increase upon zinc treatments. These NET formations of PMNs increased by either culture supernatant from PBMCs treated with zinc or rpTNF-α were inhibited by treatment of anti-rpTNF-α polyclonal antibody (pAb). These results suggested that zinc has an immunostimulating effect on the NET formation of PMNs, which is mediated by TNF-α released from zinc-treated PBMCs. Therefore, zinc may play an important role for NET formation in the defense of porcine inflammatory diseases.

• 수산해양기술연구
• /
• 제53권3호
• /
• pp.246-255
• /
• 2017

The species composition and variation in abundance of fishery resources near Oenarodo, Go-heung Peninsula, Korea, were investigated by gill net, trap net, and longline in May, July, and October 2015 and 2016. During the study period, the total catch included 14 species in the gill net, 11 species in the trap net, and 4 species in the longline. The dominant species were Portunus trituberculatus and Raja pulchrain the gill net, Charybdis japonicaand and Octopus vulgarisin in the trap net, and Muraenesox cinereusin in the longline. The Catch Per Unit Effort (CPUE) per individual and per weight in the gill net were similar in May and July of 2015 and 2016. In October 2015, the CPUE per individual was 2.1 ind./h and the CPUE per weight was 505 g/h higher than the results in 2016, but there was no significant difference in the total CPUE between 2015 and 2016. In the trap net, the CPUE per weight was similar in both 2015 and 2016, but the CPUE per individual was 2.7 ind./h higher in October 2015 than in October 2016 and the total CPUE was not significantly different from 2015 to 2016. The CPUE per individual and weight in the longline were significantly higher in July and October 2015 than in the same months of 2016, but the total CPUE in 2015 and 2016 did not show a significant difference.

• 스마트미디어저널
• /
• 제11권5호
• /
• pp.48-55
• /
• 2022

토마토 작물은 병해에 노출이 쉽고 단시간에 퍼지므로 병해에 대한 늦은 조치로 인한 피해는 생산량과 매출에 직접적인 영향을 끼친다. 따라서, 토마토의 병해에 대해 누구나 현장에서 간편하고 정확하게 진단하여 조기 예방을 가능하게 하는 서비스가 요구된다. 본 논문에서는 사전에 ImageNet 전이 학습된 딥러닝 기반 모델을 적용하여 토마토의 9가지 병해 및 정상인 경우의 클래스를 분류하고 서비스를 제공하는 시스템을 구성한다. Plant Village 데이터 셋으로부터 토마토 병해 및 정상을 분류한 잎의 이미지 셋을 합성곱을 사용하여 조금 더 가벼운 신경망을 구축한 딥러닝 기반 CNN구조를 갖는 MobileNet, ResNet의 입력을 사용한다. 2가지 제안 모델의 학습을 통해 정확도와 학습속도가 빠른 MobileNet를 사용하여 빠르고 편리한 서비스를 제공할 수 있다.