Proc. SPIE. 6149, 2nd International Symposium on Advanced Optical Manufacturing and Testing Technologies: Advanced Optical Manufacturing Technologies
KEYWORDS: Near infrared, Statistical analysis, Data modeling, Manufacturing, Inspection, Image classification, Near infrared spectroscopy, Binary data, Scalable video coding, Library classification systems
The fast qualitative analysis of textile fiber is a crucial step in textile manufacture, export and inspection. This paper presents a near infrared spectroscopy classification method based on SVM for fast qualitative analysis of textile fiber. SVM is a new automatic classification tool and it has successfully been applied to standard classification tasks, such as text classification, pattern identification, bioinformatics and medical diagnosis. In this paper, SVM is extended into near infrared fast qualitative analysis of textile fiber for the first time. In this paper, eight kinds classification algorithms which are composed of two classifiers(C-SVC and ν-SVC) and four kernel functions (linear, polynomial, RBF and sigmoid) are used to do classification experiments and comparison analysis for ten kinds familiar textiles fiber. Experiment results show that it is feasible to apply SVM in fast qualitative analysis of textile fiber, and the optimal classifier algorithm and the corresponding experimental results are reported.
The NIR region is composed of radiation with wavelengths of 700-2500nm. The analytical technology of NIR has many virtues, such as fast (one minute or less per sample), nondestructive, suitable for on-line use. So it can be applied to the textile field. But because of the interference from strongly overlapping constituents' spectra and from light scatter variations, the transformations of the diffuse spectroscopy measurements should ideally pass through two stages, response linearization and optical correction. Before being used in linear calibration model, the spectra data usually is pretreated by the different pretreatment methods. The pretreatment methods contain derivative, smoothing, normalizing, data compression and so on. These pretreatment methods resolve the overlapping peaks, remove the linear baselines and eliminate the spectral noise. Then three methods, Multiple Linear Regression (MLR), Partial Least-Squares (PLS) and Neural networks are adopted to establish a model to with the pretreated spectra data. The first two methods express a linear relationship between the spectral data and the concentration. And the third method is a nonlinear method. The validation sample set is used to validate these three established models. Depending on the comparison of the results, the best linear calibration model to estimate the unknown samples is set up.
Near Infrared (NIR) spectroscopy in the region from 1300 to 1700nm, coupled with multivariate analytic statistical techniques, have been used to predict the chemical properties of textile fiber. Molecule absorbs electromagnetic wave with especial wavelength, which leads to bring characteristic absorption spectrum. Characteristic wavelength is the most important parameter in NIR detection. How to select characteristic wavelength is the key to NIR measure. Different
mathematical methods are used to find relationship between the NIR absorption spectrum and the chemical properties of the textile fiber. We adopt stepwise multiple linear regression (SMLR) to select characteristic wavelength. As objective condition is limited, this article only refers to cotton and terylene. By computing correlation coefficient, we establish calibration equation with the smoothed absorbance data. Finally, the bias was controlled under 6%. Then, we find that NIR can be used to carry on qualitative analysis and quantitative analysis of the textile.
Photosynthesis of plants is to absorb the special wavelength of sunlight by the chlorophylls. According to the absorption spectrum of chlorophylls, we managed to make a LED lamp for the growing of green plants, and the relative energy spectrum distribution of the lamp match with the absorbing spectrum of green plants. Because the absorption wavelength range of chlorophylls are respectively 390~420nm, 430~460nm and 650~680nm, we choose different peak wavelength LEDs which are respectively at 400nm, 450nm, 655nm. By calculation, the general energy ratio of the three types of LEDs is 22:46:33, which corresponds to the absorption spectrum of chlorophylls. The illuminance of lamp for the growing of green plants on plants away 0.5 meter is 80lx by measuring. The LEDs lamp can be used to complement light and increase the efficiency of photosynthesis in cloudy, in door or at night. In another word, the photosynthesis is more powerful, and the more carbohydrates are synthesized, supplying enough material and energy for the growing of green plants.