센서 융합 데이터를 활용한 논 토양 성분 예측 = Prediction of soil properties in paddy soil using sensor fusion data
저자
발행사항
전주 : 전북대학교 일반대학원, 2023
학위논문사항
학위논문(석사)-- 전북대학교 일반대학원 : 농업기계공학(농업기계공학) 2023. 2
발행연도
2023
작성언어
한국어
주제어
발행국(도시)
전북특별자치도
형태사항
viii, 76 p. ; 26 cm
일반주기명
지도교수: 조용진
UCI식별코드
I804:45011-000000056106
소장기관
This study measured the paddy soil properties and soil sensor data in Korea and analyzed a prediction model of soil properties based on sensor data. This study gathered soil samples from four paddy fields located in Suchon-ri, Jangan-myeon, Hwaseong-si, Gyeonggi-do. Then, the properties of soil samples were investigated: soil sensors such as soil cone index (CI), which indicates soil hardness, electric conductivity (EC), and spectrum, were measured. The regression models of soil properties-spectrum were analyzed in paddy soils of Korea using stepwise multiple linear regression (SMLR) and partial least squares regression (PLSR). This study examined the fusion data of CI, EC, and spectrum measured by the soil sensor with the PLSR model and compared it with the results of the PLSR model using only the spectrum. The conclusions made in this study are as follows.
1. This study investigated the soil properties of paddy soil samples: pH, EC, Cation exchange capacity (CEC), Extractable calcium (Ca+2), Extractable magnesium (Mg+2), Extractable kalium (K+), Soil organic matter (SOM), Total nitrogen (TN), Total organic carbon (TOC), and Phosphorus pentoxide (P2O5). The mean and standard deviation of the investigated paddy soil properties were 7.87±0.37, 1.79 dS/m±0.65, 21.34 cmolc/kg±1.51, 5.79 cmolc/kg±0.70, 5.73 cmolc/kg±0.62, 1.36 cmolc/kg±0.14, 20.57 g/kg±4.48, 0.12 g/kg±0.02, 1.20%±0.26, 44.24 mg/kg±12.73, respectively. Soil sensors were used in this study: EC, CI, and spectrum. Soil EC was measured at 15 cm depth from the topsoil, and soil CI has used the average CI values calculated every 1 cm from the topsoil to 18 cm depth. The soil spectrum was measured 10 times after drying the collected soils, and the results were averaged. The average and standard deviation of the EC and CI data in paddy soil were 1,136.79 kPa±303.82 and 2.47 dS/m±0.22, respectively. EC was about 38% higher than the overall average of 2.47 dS/m in Field 1 and about 14% lower in Fields 2, 3, and 4. Compared to the average 1136.79 kPa of soil CI, Field 1 showed about a 27% decrease and increased by about 26% compared to the value of Field 4.
2. When analyzing the correlation between the investigated soil properties and sensor data, CI sensor data showed the highest correlation (r=-0.712) with soil properties of CEC, followed by Mg+2, pH, and Ca+2 in order (r=-0.673, -0.669, 0.617). EC sensor data showed the highest correlation (r=0.745) with EC property, followed by Ca+2, Mg+2, and K+ in order (r=-0.663, -0.597, 0.575, respectively).
3. Among the prediction models of soil properties based on soil spectroscopy using PLSR, the soil properties evaluated by the Fair model were pH (R2v=0.66, RPDv=1.67), CEC (R2v=0.64, RPDv=1.66), Mg+2 (R2v=0.67, RPDv=1.68) and Ca+2 (R2v=0.57, RPDv=1.53). Among the validation models of soil properties based on the PLSR prediction model, the soil property verified as a Good model was pH (RPD=2.23), and the soil properties verified as a Fair model were CEC (RPD=1.98), Ca+2 (RPD=1.67) and Mg+2 (RPD=1.98).
4. Among the prediction models of soil properties based on soil spectroscopy using SMLR, the soil properties that are evaluated as fair model were pH and Mg+2. From the results of the SMLR analysis, the significant wavelengths in the spectral band were found to include all the VIS-NIR spectral bands in TN, and significant wavelengths were investigated only in the NIR spectral band in other soil properties. When comparing the B-matrix coefficients derived from the PLSR analysis and the significant wavelengths of SMLR, some significant wavelengths were similar. This tendency was the same for soil properties with high R2 of the regression model.
5. The result of the soil composition prediction model by combining soil spectrum with CI and EC sensor data increased the value of RPDv by at least about 3% and at most about 22% compared to the results of the prediction model of soil properties based only on soil spectrum. In particular, in the case of K+ and P2O5, a spectrum-based PLSR analysis result was evaluated as a poor model. Still, the predicting model of sensor fusion data was evaluated as a Fair model. In the case of CEC, which had a high correlation with soil CI sensor data, the PLSR analysis fused with CI data showed the highest statistical value with RPDv=1.81. In the case of EC, which had a high correlation with soil EC sensor, PLSR fused with EC data, and the analysis showed a high statistical value with RPDv=1.46. The soil sensor data affected the prediction model of soil properties when analyzing the PLSR in which the soil sensor is fused. Based on the results of this study, it is more advantageous to use the PLSR analysis of sensor fusion data when developing a prediction model of soil properties.
In the predictive regression model for soil properties in this study, PLSR analysis of sensor fusion showed higher accuracy than other analysis methods. The higher the correlation between the soil properties and the sensor data, the higher the statistical value in the PLSR analysis of the sensor fusion. However, despite applying the PLSR analysis of sensor fusion, there was no Good model in the model evaluation criteria. For further studies, the reliability and accuracy of the model are expected to increase when the prediction model is designed by accumulating various paddy fields' data. It is expected that active data accumulation of soil spectra at the national, regional, and city levels will significantly improve the accuracy and stability of the prediction model of soil properties.
분석정보
서지정보 내보내기(Export)
닫기소장기관 정보
닫기권호소장정보
닫기오류접수
닫기오류 접수 확인
닫기음성서비스 신청
닫기음성서비스 신청 확인
닫기이용약관
닫기학술연구정보서비스 이용약관 (2017년 1월 1일 ~ 현재 적용)
학술연구정보서비스(이하 RISS)는 정보주체의 자유와 권리 보호를 위해 「개인정보 보호법」 및 관계 법령이 정한 바를 준수하여, 적법하게 개인정보를 처리하고 안전하게 관리하고 있습니다. 이에 「개인정보 보호법」 제30조에 따라 정보주체에게 개인정보 처리에 관한 절차 및 기준을 안내하고, 이와 관련한 고충을 신속하고 원활하게 처리할 수 있도록 하기 위하여 다음과 같이 개인정보 처리방침을 수립·공개합니다.
주요 개인정보 처리 표시(라벨링)
목 차
3년
또는 회원탈퇴시까지5년
(「전자상거래 등에서의 소비자보호에 관한3년
(「전자상거래 등에서의 소비자보호에 관한2년
이상(개인정보보호위원회 : 개인정보의 안전성 확보조치 기준)개인정보파일의 명칭 | 운영근거 / 처리목적 | 개인정보파일에 기록되는 개인정보의 항목 | 보유기간 | |
---|---|---|---|---|
학술연구정보서비스 이용자 가입정보 파일 | 한국교육학술정보원법 | 필수 | ID, 비밀번호, 성명, 생년월일, 신분(직업구분), 이메일, 소속분야, 웹진메일 수신동의 여부 | 3년 또는 탈퇴시 |
선택 | 소속기관명, 소속도서관명, 학과/부서명, 학번/직원번호, 휴대전화, 주소 |
구분 | 담당자 | 연락처 |
---|---|---|
KERIS 개인정보 보호책임자 | 정보보호본부 김태우 | - 이메일 : lsy@keris.or.kr - 전화번호 : 053-714-0439 - 팩스번호 : 053-714-0195 |
KERIS 개인정보 보호담당자 | 개인정보보호부 이상엽 | |
RISS 개인정보 보호책임자 | 대학학술본부 장금연 | - 이메일 : giltizen@keris.or.kr - 전화번호 : 053-714-0149 - 팩스번호 : 053-714-0194 |
RISS 개인정보 보호담당자 | 학술진흥부 길원진 |
자동로그아웃 안내
닫기인증오류 안내
닫기귀하께서는 휴면계정 전환 후 1년동안 회원정보 수집 및 이용에 대한
재동의를 하지 않으신 관계로 개인정보가 삭제되었습니다.
(참조 : RISS 이용약관 및 개인정보처리방침)
신규회원으로 가입하여 이용 부탁 드리며, 추가 문의는 고객센터로 연락 바랍니다.
- 기존 아이디 재사용 불가
휴면계정 안내
RISS는 [표준개인정보 보호지침]에 따라 2년을 주기로 개인정보 수집·이용에 관하여 (재)동의를 받고 있으며, (재)동의를 하지 않을 경우, 휴면계정으로 전환됩니다.
(※ 휴면계정은 원문이용 및 복사/대출 서비스를 이용할 수 없습니다.)
휴면계정으로 전환된 후 1년간 회원정보 수집·이용에 대한 재동의를 하지 않을 경우, RISS에서 자동탈퇴 및 개인정보가 삭제처리 됩니다.
고객센터 1599-3122
ARS번호+1번(회원가입 및 정보수정)