• 제목/요약/키워드: Bark Identification

검색결과 54건 처리시간 0.025초

심층 학습 모델을 이용한 수피 인식 (Bark Identification Using a Deep Learning Model)

  • 김민기
    • 한국멀티미디어학회논문지
    • /
    • 제22권10호
    • /
    • pp.1133-1141
    • /
    • 2019
  • Most of the previous studies for bark recognition have focused on the extraction of LBP-like statistical features. Deep learning approach was not well studied because of the difficulty of acquiring large volume of bark image dataset. To overcome the bark dataset problem, this study utilizes the MobileNet which was trained with the ImageNet dataset. This study proposes two approaches. One is to extract features by the pixel-wise convolution and classify the features with SVM. The other is to tune the weights of the MobileNet by flexibly freezing layers. The experimental results with two public bark datasets, BarkTex and Trunk12, show that the proposed methods are effective in bark recognition. Especially the results of the flexible tunning method outperform state-of-the-art methods. In addition, it can be applied to mobile devices because the MobileNet is compact compared to other deep learning models.

수피 특징 추출을 위한 상용 DCNN 모델의 비교와 다층 퍼셉트론을 이용한 수종 인식 (Comparison of Off-the-Shelf DCNN Models for Extracting Bark Feature and Tree Species Recognition Using Multi-layer Perceptron)

  • 김민기
    • 한국멀티미디어학회논문지
    • /
    • 제23권9호
    • /
    • pp.1155-1163
    • /
    • 2020
  • Deep learning approach is emerging as a new way to improve the accuracy of tree species identification using bark image. However, the approach has not been studied enough because it is confronted with the problem of acquiring a large volume of bark image dataset. This study solved this problem by utilizing a pretrained off-the-shelf DCNN model. It compares the discrimination power of bark features extracted by each DCNN model. Then it extracts the features by using a selected DCNN model and feeds them to a multi-layer perceptron (MLP). We found out that the ResNet50 model is effective in extracting bark features and the MLP could be trained well with the features reduced by the principal component analysis. The proposed approach gives accuracy of 99.1% and 98.4% for BarkTex and Trunk12 datasets respectively.

삼칠근(三七根)의 형태(形態)에 관한 연구(硏究) (A Study on a Morphological Identification of Notoginseng Radix)

  • 문성호;이영종
    • 대한본초학회지
    • /
    • 제23권2호
    • /
    • pp.25-31
    • /
    • 2008
  • Objectives : In order to distinguish morphological characteristics of trunk bark and root bark of Ulmus davidiana var. japonica (Rehder) Nakai and the trunk bark and root bark of Hemiptelea davidii Planchon were sampled and compared in terms of their external and internal features with flour states according to their medical use, through microscopic examination. Methods : The slice of the tested material made by paraffin section technique was colored with Safranine Malachite Green contrast methods, and the flour of it was mounted by the liquid made by the same ratio of each of glycerin, acetic acid, and water, and then observed and photographed by olymphus-BHT. Results : 1. Internal Features 1) A large parenchymatous cell was observed in the phloem of the slice of both trunk bark and root bark of Ulmi Cortex, However, both of the trunk bark and root bark of Hemipteleae Cortex did not have parenchymatous cell in the phloem; instead, stone cells including much square crystal of calcium oxalate were distributed around fiber bundle, and the parenchymatous cell included much druse crystal of calcium oxalate. 2) In both the Ulmi Cortex and Hemipteleae Cortex, rhytidome was observed in trunk bark, but not in root bark, but in the parenchymatous cell of the root bark of the Ulmi Cortex contained starch grain. 2. Flour States 1) In the flour of root bark of the Ulmi Cortex, a large parenchymatous cell was observed. However, in the flour of trunk bark and root bark of Hemipteleae Cortex, no parenchymatous eel was found; instead, stone cell including square crystal of calcium oxalate and druse crystal of calcium oxalate were observed. 2) There was no remarkable difference between the trunk bark and root bark of Hemipteleae Cortex. However, starch grain was contained in the parenchymatous cell of the root bark of Ulmi Cortex but not in the trunk bark of it. Conclusions : There were some morphological differences in external, internal, and flour parts of Ulmi Cortex and Hemipteleae Cortex. In particular, there was a morphological difference in flour states between the trunk bark and root bark of Ulmi Cortex, it is possible to use microscope to distinguish their flour states.

  • PDF

오가피(五加皮)의 형태(形態)에 관한 연구(硏究) (A Study on a Morphological Identification of Acanthopanacis Cortex)

  • 김형석;한효상;이영종
    • 대한본초학회지
    • /
    • 제23권2호
    • /
    • pp.41-49
    • /
    • 2008
  • Objectives : In order to distinguish morphological characteristics of trunk bark and root bark of Ulmus davidiana var. japonica (Rehder) Nakai and the trunk bark and root bark of Hemiptelea davidii Planchon were sampled and compared in terms of their external and internal features with flour states according to their medical use, through microscopic examination. Methods : The slice of the tested material made by paraffin section technique was colored with Safranine Malachite Green contrast methods, and the flour of it was mounted by the liquid made by the same ratio of each of glycerin, acetic acid, and water, and then observed and photographed by olympus-BHT. Results : 1. Internal Features 1) A large parenchymatous cell was observed in the phloem of the slice of both trunk bark and root bark of Ulmi Cortex. However, both of the trunk bark and root bark of Hemipteleae Cortex did not have parenchymatous cell in the phloem; instead, stone cells including much square crystal of calcium oxalate were distributed around fiber bundle, and the parenchymatous cell included much druse crystal of calcium oxalate. 2) In both the Ulmi Cortex and Hemipteleae Cortex, rhytidome was observed in trunk bark, but not in root bark, but in the parenchymatous cell of the root bark of the Ulmi Cortex contained starch grain. 2. Flour States 1) In the flour of root bark of the Ulmi Cortex, a large parenchymatous cell was observed. However, in the flour of trunk bark and root bark of Hemipteleae Cortex, no parenchymatous eel was found; instead, stone cell including square crystal of calcium oxalate and druse crystal of calcium oxalate were observed. 2) There was no remarkable difference between the trunk bark and root bark of Hemipteleae Cortex. However, starch grain was contained in the parenchymatous cell of the root bark of Ulmi Cortex but not in the trunk bark of it. Conclusions : There were some morphological differences in external, internal, and flour parts of Ulmi Cortex and Hemipteleae Cortex. In particular, there was a morphological difference in flour states between the trunk bark and root bark of Ulmi Cortex, it is possible to use microscope to distinguish their flour states.

  • PDF

수목 동정을 위한 수피 분류 데이터셋 구축과 합성곱 신경망 기반 53개 수종의 동정 모델 개발 (Construction of a Bark Dataset for Automatic Tree Identification and Developing a Convolutional Neural Network-based Tree Species Identification Model)

  • 김태경;백규헌;김현석
    • 한국산림과학회지
    • /
    • 제110권2호
    • /
    • pp.155-164
    • /
    • 2021
  • 자연환경에 대한 국민들의 관심 증가로 스마트폰과 같은 휴대용 기기를 이용한 수목 동정의 자동화에 대한 요구가 증가하고 있다. 최근 딥러닝 기술의 발전에 힘입어, 외국에서는 수목 인식 분야에의 적용이 활발하게 이루어지고 있다. 수목의 분류를 위해 꽃, 잎 등 다양한 형질들을 대상으로 연구가 진행되고 있지만, 접근성을 비롯한 여러 장점을 가진 수피의 경우 복잡도가 높고 자료가 부족하여 연구가 제한적이었다. 본 연구에서는 국내에서 흔히 관찰 가능한 수목 54종의 사진자료를 약 7,000 여장 수집 및 공개하였고, 이를 해외의 20 수종에 대한 BarkNet 1.0의 자료와 결합하여 학습에 충분한 수의 사진 수를 가지는 53종을 선정하고, 사진들을 7:3의 비율로 나누어 훈련과 평가에 활용하였다. 분류 모델의 경우, 딥러닝 기법의 일종인 합성곱 신경망을 활용하였는데, 가장 널리 쓰이는 VGGNet (Visual Geometry Group Network) 16층, 19층 모델 두 가지를 학습시키고 성능을 비교하였다. 또한 본 모형의 활용성 및 한계점을 확인하기 위하여 학습에 사용하지 않은 수종과 덩굴식물과 같은 방해 요소가 있는 사진들에 대한 모델의 정확도를 확인하였다. 학습 결과 VGG16과 VGG19는 각각 90.41%와 92.62%의 높은 정확도를 보였으며, 더 복잡도가 높은 모델인 VGG19가 조금 더 나은 성능을 보임을 확인하였다. 학습에 활용되지 않은 수목을 동정한 결과 80% 이상의 경우에서 같은 속 또는 같은 과에 속한 수종으로 예측하는 것으로 드러났다. 반면, 이끼, 만경식물, 옹이 등의 방해 요소가 존재할 경우 방해요소가 자치하는 비중에 따라 정확도가 떨어지는 것이 확인되어 실제 현장에서 이를 보완하기 위한 방법들을 제안하였다.

육계(Cortex of Cinnamomum cassia J.Presl) 추출물로부터 부정근 형성 유도물질 분리 및 동정 (Isolation and Identification of Adventitious Root Formation Inducing Substances from Cortex of cinnamomum cassia J.Presl)

  • 여주호;백정규;박지성;김건우
    • 한국자원식물학회지
    • /
    • 제37권1호
    • /
    • pp.11-21
    • /
    • 2024
  • 본 연구에서 육계 추출물로부터 cinnamyl alcohol을 분리·정제하였다. 연구 결과, H3 분획100 ㎍/mL에서 부정근 형성율이 286.6%였으며, 표준품 cinnamyl alcohol 1 mM (134.2 ㎍/mL)의 형성율이 290%로 유사한 결과를 나타내었다. 따라서 육계 methanol 추출물에 함유되어있는 cinnamyl alcohol이 녹두의 부정근 형성을 유도하는 주된 화합물일 것으로 판단된다. 육계에서 동정한 cinnamyl alcohol은 현재 식품첨가제 및 향료로써 사용되고 있으므로 안전성이 이미 보장되어 있으며, 한약재로써 사용된 육계 잔류물을 활용하여 부정근 형성 유도물질을 추출한다면비용적인 측면에서 또한 긍정적인 요인으로 작용할 것으로 판단된다. 따라서 육계는 유기농 농자재로써 충분히 사용 가능할 뿐만 아니라 신규 천연 식물생장조절제의 개발용 소재로서도 활용 가능할 것으로 사료된다.

잣나무 수피의 Stilbene glycosides의 분리 및 동정 (Isolation and Identification of Stilbene glycosides from the Bark of Pinus koraiensis)

  • 송홍근
    • Journal of the Korean Wood Science and Technology
    • /
    • 제29권4호
    • /
    • pp.97-102
    • /
    • 2001
  • 약 25년 생 잣나무 수피 추출물중 EtOAc용해분에 함유된 화합물을 단리 동정하였다. 화합물의 단리는 Sephadex-LH20과 TSK-gel HW-40F를 충전제로 한 Column chromatography로 하였다. 단리된 화합물의 구조동정은 $^1H{\cdot}^{13}C$-NMR, HMQC, HMBC등과 $FAB^+$ MS 등을 이용하여 결정하였다. 새로이 동정된 화합물은 Z-pinostilbenoside, E-desoxyrhaponticin, 그리고 E-resveratroloside 등이었다.

  • PDF

Simultaneous Analysis of the Coloring Compounds in Indigo, Phellodendron bark, and Madder Dye Using HPLC-DAD-MS

  • Ahn, Cheunsoon;Zeng, Xia;Obendorf, S. Kay
    • 한국의류학회지
    • /
    • 제37권6호
    • /
    • pp.827-836
    • /
    • 2013
  • Indigotin, indirubin, berberine, palmatine, alizarin, and purpurin are major pigments of indigo plant, Phellodendron bark, and madder. The six pigments were examined using the HPLC-DAD-MS instrument for the purpose of the simultaneous detection of the pigments in a single sample run. The HPLC-DAD-MS method examined the individual pigment solutions in DMSO, a solution containing 6 pigments, and the DMSO extract of the silk dyed with a dye solution of 5 pigments excluding indirubin. The retention times of the HPLC chromatograms, ${\lambda}_{max}$ of the uv-vis absorption bands in the DAD analyses, and the molecular ions detected for the compound peaks in the MSD analyses were consistent throughout the analyses of individual pigment solutions, mixed pigment solutions, and dye extracted from silk dyeing. The developed instrumental method of the simultaneous detection of six pigments can identify dye in an exhumed textile if the textile is dyed using any one (or multiple) pigments of indigo, Phellodendron bark, or madder plant.

전주 마전유적 출토 목관재의 수종식별 및 연륜연대 분석 (Species Identification and Tree-Ring Dating of Coffin Woods Excavated at Ma-Jeon Relic in Jeonju, Korea)

  • 박원규;윤두형;박수현
    • Journal of the Korean Wood Science and Technology
    • /
    • 제34권6호
    • /
    • pp.12-20
    • /
    • 2006
  • 전주 마전 유적에서 출토된 회곽묘 1기에 사용된 목재의 수종 식별과 연륜연대 분석을 하였다. 수종 식별 결과, 관재는 모두 소나무속중 소나무류 (적송류)로 동정되었다. 7개의 관재로부터 총 24점의 관편에 대한 연륜연대 분석결과, 6개에 대한 연대를 부여할 수 있었다. 횡대에 쓰인 관재는 수피를 그대로 간직하고 있고 마지막 나이테가 만재까지 완전히 형성되어 있어 1637년 가을에서 1638년 봄 사이에 벌채된 목재로 관을 제작하였음을 알 수 있었다. 다만 벌채후 건조, 저장 기간으로 인하여 연륜연대와 관의 제작 및 매장 연도와는 다소 차이가 있을 수 있다.