Influence of Digital Music on Chemical Properties in Red Leaf Lettuce

Ye-Seong Kang,Seong-Heon Kim,Chan-Seok Ryu 경상대학교 농업생명과학연구원 2016 농업생명과학연구 Vol.50 No.5

The purpose of this study was to investigate alteration of chemical properties in red leaf lettuces(Lactuca sativa L.) exposed to digital music every day. The red leaf lettuces were cultivated in two hydroponic systems composed of two layers. In the first experiment, the red leaf lettuces with treatment were exposed to the digital music, while the lettuces under control condition were not exposed to the digital music. At harvest(6 weeks after planting), fresh weight and chlorophyll content were measured and compared the treatment with the control group. Subsequently, red leaf lettuces of the next experiment were compared to fresh weight, chlorophyll, ascorbic acid and anthocyanin content during different vegetation growth stages(4 weeks and 6 weeks after planting). The comparison of data for all experiment was also divided into upper and lower parts because of the difference of temperature in hydroponic systems. As a results, fresh weight and anthocyanin of the red leaf lettuces might be influenced by the difference of temperature variations. Chlorophyll of the red leaf lettuces was not easily influenced by digital music and difference of temperature. It was also shown that ascorbic acid as inactive molecule was not easily influenced by physical response like music.

Improved Plant Image Segmentation Method using Vegetation Indices and Automatic Thresholds

Seong-Heon Kim,Chan-Seok Ryu,Ye-Seong Kang,Young-Bong Min 경상대학교 농업생명과학연구원 2015 농업생명과학연구 Vol.49 No.5

We address improved plant image segmentation based on histograms which requires using a vegetation index and threshold. Image segmentation is the most important step for extracting targets, such as vegetation, from images; this affects successful detection of plant information. Forty-two field images were acquired from a soybean field using an RGB camera. Through K-means clustering analysis, we built a new vegetation index        and generated gray-scale images. Otsu and Triangle thresholds were used to convert contrast images to binary. Optimal threshold values were generally located between the Otsu and Triangle threshold values. The combined threshold method shows 98.79% and 0.95% of mean accuracy and standard deviation, respectively, whereas the Otsu and Triangle method results show 98.17±1.71% and 97.85±1.87%, respectively. These results show that the combined method has significant segmentation potential through one-way ANOVA. Then we compared the results with K-means clustering using two-sample t-test. The K-means method’s mean accuracy is 98.18±1.79%, with no significant difference between the proposed and K-means methods. However, the proposed method’s processing time is 0.60±0.01 s, i.e., twice faster than the K-means method (1.72±0.24 s).

Yield Prediction of Chinese Cabbage (Brassicaceae) Using Broadband Multispectral Imagery Mounted Unmanned Aerial System in the Air and Narrowband Hyperspectral Imagery on the Ground

( Ye Seong Kang ),( Chan Seok Ryu ),( Seong Heon Kim ),( Sae Rom Jun ),( Si Hyeong Jang ),( Jun Woo Park ),( Tapash Kumar Sarkar ),( Hye Young Song ) 한국농업기계학회 2018 바이오시스템공학 Vol.43 No.2

Purpose: A narrowband hyperspectral imaging sensor of high-dimensional spectral bands is advantageous for identifying the reflectance by selecting the significant spectral bands for predicting crop yield over the broadband multispectral imaging sensor for each wavelength range of the crop canopy. The images acquired by each imaging sensor were used to develop the models for predicting the Chinese cabbage yield. Methods: The models for predicting the Chinese cabbage (Brassica campestris L.) yield, with multispectral images based on unmanned aerial vehicle (UAV), were developed by simple linear regression (SLR) using vegetation indices, and forward stepwise multiple linear regression (MLR) using four spectral bands. The model with hyperspectral images based on the ground were developed using forward stepwise MLR from the significant spectral bands selected by dimension reduction methods based on a partial least squares regression (PLSR) model of high precision and accuracy. Results: The SLR model by the multispectral image cannot predict the yield well because of its low sensitivity in high fresh weight. Despite improved sensitivity in high fresh weight of the MLR model, its precision and accuracy was unsuitable for predicting the yield as its R² is 0.697, root-mean-square error (RMSE) is 1170 g/plant, relative error (RE) is 67.1%. When selecting the significant spectral bands for predicting the yield using hyperspectral images, the MLR model using four spectral bands show high precision and accuracy, with 0.891 for R2, 616 g/plant for the RMSE, and 35.3% for the RE. Conclusions: Little difference was observed in the precision and accuracy of the PLSR model of 0.896 for R², 576.7 g/plant for the RMSE, and 33.1% for the RE, compared with the MLR model. I f the multispectral imaging sensor composed of the significant spectral bands is produced, the crop yield of a wide area can be predicted using a UAV.

Yield Prediction of Chinese Cabbage (Brassica rapa var. glabra Regel.) using Narrowband Hyperspectral Imagery and Effective Accumulated Temperature

Ye-Seong Kang,Sae-Rom Jun,Si-Hyeong Jang,Jun-Woo Park,Hye-Young Song,Chan-Seok Ryu 경상대학교 농업생명과학연구원 2020 농업생명과학연구 Vol.54 No.3

In this paper, the model for predicting yields of chinese cabbages of each cultivar (joined-up in 2015 and wrapped-up in 2016) was developed after the reflectance of hyperspectral imagery was merged as 10 nm, 25 nm and 50 nm of FWHM (full width at half maximum). Band rationing was employed to minimize the unstable reflectance of multi-temporal hyperspectral imagery. The stepwise analysis was employed to select key band ratios to predict yields in all cultivars. The key band ratios selected for each of FWHM were used to develop the yield prediction models of chinese cabbage for all cultivars (joined-up & wrapped-up) and each cultivar (joined-up, wrapped-up). Effective accumulated temperature (EAT) was added in the models to evaluate its improvement of performances. In all models, the performance of models was improved with adding of EAT. The models with EAT for each of FWHM showed the predictability of yields in all cultivars as R2≥0.80, RMSE≤694 g/plant and RE≤28.3%. Such as this result, if the yield can be predicted regardless of the cultivar, it is considered to be advantageous when predicting the yield over a wide area because it is not require a cultivar classification work as pre-processing in imagery.

Estimating Moisture Content of Cucumber Seedling Using Hyperspectral Imagery

Kang, Jeong-Gyun,Ryu, Chan-Seok,Kim, Seong-Heon,Kang, Ye-Seong,Sarkar, Tapash Kumar,Kang, Dong-Hyeon,Kim, Dong Eok,Ku, Yang-Gyu Korean Society for Agricultural Machinery 2016 바이오시스템공학 Vol.41 No.3

Purpose: This experiment was conducted to detect water stress in terms of the moisture content of cucumber seedlings under water stress condition using a hyperspectral image acquisition system, linear regression analysis, and partial least square regression (PLSR) to achieve a non-destructive measurement procedure. Methods: Changes in the reflectance spectrum of cucumber seedlings under water stress were measured using hyperspectral imaging techniques. A model for estimating moisture content of cucumber seedlings was constructed through a linear regression analysis that used the moisture content of cucumber seedlings and a normalized difference vegetation index (NDVI). A model using PLSR that used the moisture content of cucumber seedlings and reflectance spectrum was also created. Results: In the early stages of water stress, cucumber seedlings recovered completely when sub-irrigation was applied. However, the seedlings suffering from initial wilting did not recover when more than 42 h passed without irrigation. The reflectance spectrum of seedlings under water stress decreased gradually, but increased when irrigation was provided, except for the seedlings that had permanently wilted. From the results of the linear regression analysis using the NDVI, the model excluding wilted seedlings with less than 20% (n=97) moisture content showed a precision ($R^2$ and $R^2_{\alpha}$) of 0.573 and 0.568, respectively, and accuracy (RE) of 4.138% and 4.138%, which was higher than that for models including all seedlings (n=100). For PLS regression analysis using the reflectance spectrum, both models were found to have strong precision ($R^2$) with a rating of 0.822, but accuracy (RMSE and RE) was higher in the model excluding wilted seedlings as 5.544% and 13.65% respectively. Conclusions: The estimation model of the moisture content of cucumber seedlings showed better results in the PLSR analysis using reflectance spectrum than the linear regression analysis using NDVI.

Model Assessment Multi-temporal Monitoring of Chinese Cabbage Growth using Low Altitude Remote Sensing System

Ye Seong Kang,Seong Heon Kim,Jeong Gyun Kang,Tapash Kumar Sarkar,Young Seok Kwon,Sae Rom Jun,Won Jun Kim,Chan Seok Ryu 경상대학교 농업생명과학연구원 2017 농업생명과학연구 Vol.51 No.4

It is necessary to monitor growth status of the crops due to continuous change of climate causing insecurity in crop cultivation. Low altitude remote sensing(LARS) system is utilized to accurately monitor the growth status of the crops. In this study, models for monitoring fresh weight(FW), one of the major growth factors of Chinese cabbage, were developed with structural indices and simple ratio calculated from bands in remotely sensed canopies by NIR, RE(imaging sensor A) and multispec-4c sensors(imaging sensor B) equipped with fixed-wing UAV depending on vegetation stages of normal planting(NP) and delayed planting(DP) Chinese cabbages. In results of imaging sensor A, the estimation models using structural indices and simple ratio were divided into NP and DP due to different attribute of reflectance in canopies with changed environment condition depending on different planting dates. The estimation models using simple ratio calculated by red edge and visible bands of NP showed better performance than other models, but RMSE was high. The models using simple ratio calculated by same bands of DP were feasible to accurately estimate FW(R2 of more than 0.946 with RMSE of less than 169.5 g). In results of imaging sensor B, the estimation models using structural indices and simple ratio on DP were divided into low to intermediate FW and intermediate to high FW. As a result, estimation models of all structural indices and simple ratio in low to intermediate FW bands were advisable to estimate FW(R2 of more than 0.860 with RMSE of less than 104.7 g). Estimation models of those calculated by red edge and visible bands in intermediate to high FW were only possible to accurately estimate FW(R2 of more than 0.532 with RMSE of less than 400.7 g).

Yield Prediction of Chinese Cabbage (Brassicaceae) Using Broadband Multispectral Imagery Mounted Unmanned Aerial System in the Air and Narrowband Hyperspectral Imagery on the Ground

Kang, Ye Seong,Ryu, Chan Seok,Kim, Seong Heon,Jun, Sae Rom,Jang, Si Hyeong,Park, Jun Woo,Sarkar, Tapash Kumar,Song, Hye young Korean Society for Agricultural Machinery 2018 바이오시스템공학 Vol.43 No.2

Purpose: A narrowband hyperspectral imaging sensor of high-dimensional spectral bands is advantageous for identifying the reflectance by selecting the significant spectral bands for predicting crop yield over the broadband multispectral imaging sensor for each wavelength range of the crop canopy. The images acquired by each imaging sensor were used to develop the models for predicting the Chinese cabbage yield. Methods: The models for predicting the Chinese cabbage (Brassica campestris L.) yield, with multispectral images based on unmanned aerial vehicle (UAV), were developed by simple linear regression (SLR) using vegetation indices, and forward stepwise multiple linear regression (MLR) using four spectral bands. The model with hyperspectral images based on the ground were developed using forward stepwise MLR from the significant spectral bands selected by dimension reduction methods based on a partial least squares regression (PLSR) model of high precision and accuracy. Results: The SLR model by the multispectral image cannot predict the yield well because of its low sensitivity in high fresh weight. Despite improved sensitivity in high fresh weight of the MLR model, its precision and accuracy was unsuitable for predicting the yield as its $R^2$ is 0.697, root-mean-square error (RMSE) is 1170 g/plant, relative error (RE) is 67.1%. When selecting the significant spectral bands for predicting the yield using hyperspectral images, the MLR model using four spectral bands show high precision and accuracy, with 0.891 for $R^2$, 616 g/plant for the RMSE, and 35.3% for the RE. Conclusions: Little difference was observed in the precision and accuracy of the PLSR model of 0.896 for $R^2$, 576.7 g/plant for the RMSE, and 33.1% for the RE, compared with the MLR model. If the multispectral imaging sensor composed of the significant spectral bands is produced, the crop yield of a wide area can be predicted using a UAV.

초분광영상 이용 오이 및 수박 묘의 수분함량 추정

김성헌(Seong-Heon Kim),강정균(Jeong-Gyun Kang),유찬석(Chan-Seok Ryu),강예성(Ye-Seong Kang),Tapash Kumar Sarkar,강동현(Dong Hyeon Kang),구양규(Yang-Gyu Ku),김동억(Dong-Eok Kim) (사)한국생물환경조절학회 2018 생물환경조절학회지 Vol.27 No.1

본 연구는 초분광 영상을 이용하여 오이 및 수박과 같은 박과 묘의 수분함량을 추정하기 위해 수행되었다. 오이와 수박 묘 샘플에 수분 스트레스를 가한 후 초분광 영상 취득 시스템을 이용하여 오이와 수박 묘 잎을 촬영하여 반사율을 계산하였고, 건조기를 이용하여 해당 모종의 수분함량을 측정하였다. 마지막으로 영상의 반사율과 수분함량을 이용하여 부분최소제곱회귀분석을 통해 수분함량 추정모델을 개발하였다. 오이 묘 수분함량 추정모델은 R² 0.73, RMSE 1.45%, RE 1.58%의 성능을 보였으며, 수박 묘 수분함량 추정모델은 R² 0.66, RMSE 1.06%, RE 1.14%의 성능을 보였다. 유효범위를 넘어가는 극단치를 제거하여 모델의 성능을 다시 분석한 결과, 오이 모델의 경우 R² 0.79, RMSE 1.10%, RE 1.20으로 상승하였다. 오이와 수박 묘를 함께 분석하여 모델을 제작한 결과, R² 0.67, RMSE 1.26, RE 1.36으로 분석되었다. 오이 모델이 수박 모델보다 비교적 높은 성능을 보였는데, 이러한 원인은 오이의 수분함량 변이가 넓게 분포되어 있었기 때문이라고 판단된다. 또한 데이터셋에서 유효범위를 넘어가는 극단치를 제거한 결과 오이 모델의 정확도 및 정밀도가 상승하였다. 결론적으로 오이 및 수박 묘 수분함량 추정모델들의 추정선의 기울기 차가 크지 않고, 서로 교차되기 때문에 두 모델들은 모두 수분함량을 추정하는데 있어서 유의한 것으로 판단된다. 또한 샘플의 변수가 넓게 분포된 변이를 갖는다면 추정모델의 정확도와 정밀도는 분명 상승할 것이며, 개선된 모델을 이용하면 저가형 센서를 개발하는데 활용될 수 있을 것으로 사료된다. This research was conducted to estimate moisture content in cucurbitaceae seedlings, such as cucumber and watermelon, using hyperspectral imagery. Using a hyperspectral image acquisition system, the reflectance of leaf area of cucumber and watermelon seedlings was calculated after providing water stress. Then, moisture content in each seedling was measured by using a dry oven. Finally, using reflectance and moisture content, the moisture content estimation models were developed by PLSR analysis. After developing the estimation models, performance of the cucumber showed 0.73 of R², 1.45% of RMSE, and 1.58% of RE. Performance of the watermelon showed 0.66 of R², 1.06% of RMSE, and 1.14% of RE. The model performed slightly better after removing one sample from cucumber seedlings as outlier and unnecessary. Hence, the performance of new model for cucumber seedlings showed 0.79 of R², 1.10% of RMSE, and 1.20% of RE. The model performance combined with all samples showed 0.67 of R², 1.26% of RMSE, and 1.36% of RE. The model of cucumber showed better performance than the model of watermelon. This is because variables of cucumber are consisted of widely distributed variation, and it affected the performance. Further, accuracy and precision of the cucumber model were increased when an insignificant sample was eliminated from the dataset. Finally, it is considered that both models can be significantly used to estimate moisture content, as gradients of trend line are almost same and intersected. It is considered that the accuracy and precision of the estimating models possibly can be improved, if the models are constructed by using variables with widely distributed variation. The improved models will be utilized as the basis for developing low-priced sensors.

고정익 무인기로 획득한 다중분광 영상을 이용한 무의 생육 추정

강예성 ( Ye Seong Kang ),김성헌 ( Seong Heon Kim ),강정균 ( Jeong Gyun Kang ),전새롬 ( Sae Rom Jun ),김원준 ( Won Jun Kim ),타파스쿠마 ( Tapash Kumar Sarkar ),유찬석 ( Chan Seok Ryu ) 한국농업기계학회 2016 한국농업기계학회 학술발표논문집 Vol.21 No.2

본 연구는 고정익 무인기(eBee, Sensefly, Swiss)에 탑재된 다중분광센서(Multispec4C, Airinov, Switzerland)로 취득한 무 포장의 영상(공간분해능:6cm)을 이용하여 산출한 식생지수로 잎 생체중과 수확량을 추정하기 위해 수행되었다. 다중분광센서의 Green, Red, Red edge, NIR 파장영역을 이용하여 무의 캐노피를 생육단계별로 촬영하였고 5개 무의 잎에 평균 반사값으로 NDVI와 개체영역을 지정하면서 생길 수 있는 토양의 영향을 최소화하기 SAVI 및 OSAVI를 각각 산출하였다. 산출된 식생지수(NDVI, SAVI, OSAVI)와 측정한 생육 데이터를 이용하여 상관 및 회귀분석을 하였다. 무의 수확량 추정모델을 분석한 결과 무의 생육 기간이 길어짐(50일 이후)에 따라 NDVI, SAVI 및 OSAVI값이 포화되는 경향이 나타나 모델의 정확도 및 정밀도가 낮아졌다. 따라서 생육기간이 50일 이하인 데이터를 이용하여 다시 분석한 결과 NDVI를 이용한 모델에서는 정확도와 정밀도가 높아졌으나 SAVI 및 OSAVI의 결과보다는 낮았다. SAVI를 이용한 추정모델은 정확도(R2)가 0.718, 정밀도(RMSE)가 162.4g로 나타났고, OSAVI를 이용한 추정모델은 정확도(R2)가 0.653, 정밀도(RMSE)가 180.1g로 나타났다. 무 잎의 생체중은 무의 수확량과 높은 상관성 r=0.997을 보였기 때문에 무의 수확량 추정모델과 비슷한 경향을 보였다. 결과적으로 무의 수확량과 무 잎의 생체중을 추정하기 위한 모델에서 생육기간에 따라 큰 성능차이를 보였고 그 중 식생지수 SAVI를 이용한 모델이 다른 식생지수를 이용한 모델보다 개체영역별 무의 생육을 추정하기에 가장 좋은 모델의 결과를 보였다.

지상 및 항공 초분광 이미지를 이용한 무와 배추 재배면적 예측을 위한 주요 파장 선정

강예성 ( Ye Seong Kang ),전새롬 ( Sae Rom Jun ),박준우 ( Jun Woo Park ),송혜영 ( Hye Young Song ),장시형 ( Si Hyeong Jang ),유찬석 ( Chan Seok Ryu ) 한국농업기계학회 2019 한국농업기계학회 학술발표논문집 Vol.24 No.2

본 연구에서는 지상에서 취득된 무와 배추의 초분광 이미지를 의사결정트리법으로 분석하여 생육 시기에 따라 무와 배추를 분류할 수 있는 주요 파장을 선정하였다. 선택된 주요 파장을 항공 초분광 이미지에 적용하여 무와 배추의 분광특성 차이를 이용하여 재배면적을 산출하고 실제로 조사된 재배면적과 비교하였다. 지상 초분광 이미지는 2015년 전라남도 무안군의 무와 배추 포장에서 11월 4일에 Specim PS (SPECIM, Finland)를 이용하여 2 m 높이에서 400 ~ 1000 nm를 분광 해상도 5.2 nm로 취득되었다. 항공 초분광 이미지는 2014년 10월 29일에 전라북도 고창군 대산면을 1500m 고도에서 CASI-1500 (ITRES, Canada)로 동일한 파장범위를 분광 해상도 28.8 nm로 취득되었다. 지상 초분광 이미지의 FWHM 5.2 nm는 항공 초분광 이미지의 FWHM과 비교적 넓은 FWHM으로 구성된 소형 다중분광 이미지 센서 개발을 고려하여 FWHM 25 nm로 평준화하였다. 높은 공간 해상도를 가진 지상 초분광 이미지를 이용하여 무와 배추를 분류하기 위한 의사결정트리 (학습 30%:검증 70%)를 수행한 결과, 중심 파장이 715 nm인 red edge (RE) 영역만이 선택되었고 분류 정확도로 overall accuracy (OA)는 94.7%와 kappa coefficient (KC)는 85.4%였다. 동일한 방법으로 항공 초분광 이미지를 분류한 결과에서도 RE영역에서 중심 파장 701 nm 만 선정되었고 OA는 87.8%와 KC 70.5%였다. 중심 파장이 715 nm 이고 FWHM이 25 nm 인 센서로 항공기 이미지에서 재배면적을 예측한 결과 무 56.2 ha와 배추 68.2 ha로 지상에서 조사된 무 56.3 ha와 배추 68.1 ha의 면적과 0.1 ha 차이였지만 무와 배추 포장의 분류정확도는 각각 80.7% 및 84.7%로 나타났다.