Spectroscopy

Design and verification of improved factor number selection process for parallel factor algorithm
Liu Wenya, Tian Zhaoshuo, Cui Zihao, Bi Zongjie, Fu Shiyou
2021, 50(S2): 20210362. doi: 10.3788/IRLA20210362
[Abstract](306) [FullText HTML] (127) [PDF 1268KB](19)
In order to solve the problem that the selection process of the number selection of model factors in the decomposition of three-dimensional fluorescence spectrum by parallel factor algorithm is not clear, an improved factor number selection process composed of core diagonal matrix, kernel uniform function and constant wavelength residual graph was proposed. The improved parallel factor analysis algorithm was developed to verify the accuracy of factor number selection process with humic acid as detection material. The results show that, combined with the above process, when the excitation light and emission light are in 350-450 nm/350-620 nm, respectively, and the factor number is 4, the core diagonal matrix distribution meets the demand, the kernel consistent function is 52%, the residual error of the fitting diagram is the smallest, and the decomposition effect is the best in the standard region. Compared with using a single method, the above combination process is more logical and accurate, and can quickly determine the number of factors in practical application. The four factors are two humic acid factor A located at 360-370 nm/450-500 nm and 350-360 nm/450-500 nm, one humic acid factor C located at 365-375 nm/475-525 nm, and one soil fulvic acid factor located at 380-390 nm/475-525 nm. When the concentration increased from 20 mg/L to 200 mg/L, the composition and contribution rate of the factors has little difference, that is, the change of concentration did not change the properties of the solution.
Crop classification of modern agricultural park based on time-series Sentinel-2 images
Zhang Dongyan, Dai Zhen, Xu Xingang, Yang Guijun, Meng Yang, Feng Haikuan, Hong Qi, Jiang Fei
2021, 50(5): 20200318. doi: 10.3788/IRLA20200318
[Abstract](446) [FullText HTML] (199) [PDF 1339KB](65)
Quickly and accurately grasping the spatial distribution of crops, estimating the area and scope of different crops were of great significance for the country to formulate macroscopic agricultural policies and guide farmers in agricultural production. To explore an efficient and accurate crop classification method, this paper took the agricultural area of Jalaid Banner of Hinggan League in Inner Mongolia Autonomous Region of China as the study area and extracted main crop classification based on the Sentinel-2 satellite remote sensing image data from May to October 2019. By analyzing the time-series curves of the four characteristic indexes of NDVI, RVI, EVI and Ref (NIR) in the study area, a total of four classification methods including Random Forest (RF), Decision Tree (DT), Support Vector Machine (SVM), and Maximum Likelihood (ML) were used to classify various crops in the study area. The RF results were compared with the classification results of DT, SVM and ML, and the spatial distribution of major crops such as rice, corn, stevia, dry rice and soybean were successfully extracted and identified. The results showed that RF had the highest overall classification accuracy of 95.8% with a Kappa coefficient of 0.944, DT, SVM and ML had classification accuracy of 92.2%, 91.6% and 86.5%, respectively. The above results indicate that the multitemporal Sentinel-2 remote sensing images can be extracted by spectral index time-varying features, and the crop classification using the random forest algorithm can obtain high accuracy results, which provides effective technical support for the fine guidance of large-scale agricultural production in the park.
Classification of iron ore based on machine learning and laser induced breakdown spectroscopy
Yang Yanwei, Zhang Lili, Hao Xiaojian, Zhang Ruizhong
2021, 50(5): 20200490. doi: 10.3788/IRLA20200490
[Abstract](646) [FullText HTML] (152) [PDF 1368KB](97)
Iron ore is a very important mineral resource. Its development and utilization have a great impact on the development of the iron and steel industry. The selection and classification of iron ore is an indispensable link in the metallurgical industry. Different types of iron ores and its grade will directly affect the ratio of other substances, so the research on the selection and classification of iron ore is of great significance in the metallurgical industry. Laser-induced breakdown spectroscopy (LIBS) is a recently developed component detection technology. It has the advantages of non-destructive, fast, in-situ online detection, etc., and has certain advantages in the field of chemical composition detection and sample classification. In order to study the method of improving the classification accuracy of iron ores, 10 kinds of natural iron ores, including hematite, limonite, siderite, mica hematite, magnetite, maghmite, oolitic hematite, pyrite, cobalt-bearing magnetite, pyrrhotine, were classified with LIBS and machine study. In this study, 10 kinds of natural iron ores, were ablated by LIBS to obtain their corresponding spectral data; then the 10 features corresponding to the maximum spectral intensity were obtained by setting a threshold; the classification training and testing on selected feature spectra were performed with KNN, RF, and SVM models. The results show that the classification accuracy of the three machine learning models: KNN, RF and SVM are 83.0%, 80.7%, and 90.3%, respectively. It can be seen from the classification accuracy that combination of LIBS and machine learning can achieve rapid and accurate classification of iron ores, which will provide a new method for classification of iron ores in the metallurgical industry.
Research on classification and recognition system based on miniaturized portable spectral imaging technology
Zhang Chen, Liu Shuyang, Zhao Anna, Wang Tianhe, Jia Xiaodong
2019, 48(10): 1023001. doi: 10.3788/IRLA201948.1023001
[Abstract](702) [PDF 1662KB](75)
Miniaturized portable spectral imaging was increasingly used in daily life, providing more convenience for people's lives. Citrus was one of the fruits that people often eat and store in their daily life. In the smart refrigerator, citrus was not conducive to identification and classification due to similar varieties. Spectral imaging technology used its characteristic wavelengths to realize its recognition. A new type of single-chip spectral imaging chip was used to build a compact and portable spectral imaging system. Spectral recognition technology was used to realize the classification and identification of citrus fruits, and cross-validation was carried out by batch samples, simultaneously establish the relationship between spectral resolution and accuracy, by constraining the spectral resolution, which effectively improved the classification and recognition accuracy of citrus fruits. Spectral resolution was less than 20 nm, and recognition accuracy could reach more than 95%.