Hyperspectral imaging spectrometer combines imaging technology and spectroscopy technology to detect the spatial characteristics of objects while forming a continuous spectral coverage of tens to hundreds of bands with a bandwidth of about 10 nm for each spatial pixel dispersion. It acquires hyperspectral images of scenery or targets with high spectral resolution. It is widely used in research and observation in the fields of land, atmosphere and ocean.
Hyperspectral Hyperspectral Imaging Spectrometer-Overview
The hyperspectral hyperspectral imaging spectrometer was developed on the basis of multispectral remote sensing imaging technology in the 1980s. It acquires hyperspectral images of scenery or targets with high spectral resolution, and performs land and atmosphere on aviation and spacecraft. , Ocean and other observations have a wide range of applications, hyperspectral imaging spectrometer can be applied to the precise classification of features, feature recognition, feature information extraction. After establishing the target hyperspectral remote sensing information processing and quantitative analysis model, it can improve the level of automation and intelligence of hyperspectral data processing. Due to the great advantage of the high spectral resolution of the hyperspectral imaging spectrometer, acquiring many continuous wave band spectral images while observing the ground in space, to achieve the purpose of directly identifying the surface Material of the earth from space, has become a hot spot in the field of remote sensing. Become the main technical means of contemporary space earth observation. The use of hyperspectral imaging spectrometer on the ground has also achieved great results, such as scientific research, environmental protection of industry, agriculture and forestry.
The main performance parameters of the hyperspectral imaging spectrometer are: (1) Noise equivalent reflectance difference (NEΔp), which is reflected in the signal-to-noise ratio (SNR); (2) Instantaneous field of view (IFOV), which is reflected in the ground resolution; (3 ) Spectral resolution is intuitively expressed as the number of bands and the width of the band.
The analysis and processing of remote sensing information with high spectral resolution focuses on the development and quantitative analysis of image information on the spectral dimension. The key technologies of its image processing mode are: (1) Display of hyper-multidimensional spectral image information, such as image cubes (see figure 1) Generation; ⑵ Spectral reconstruction, that is, imaging spectral data calibration, quantification, and atmospheric correction models and algorithms, to achieve image-spectral conversion of imaging spectral information; ⑶ spectral coding, especially referring to the spectral absorption position and depth , Symmetry and other spectral characteristic parameters; ⑷ ground spectral matching identification algorithm based on spectral database; ⑸ hybrid spectral decomposition model; ⑹ surface biophysical and chemical processes and parameters identification and inversion algorithm based on spectral model.
The high-resolution spectral imaging remote sensing originated from the study of geological mineral identification and mapping, and gradually expanded into the research of vegetation ecology, ocean coast color, ice, snow, soil and atmosphere.
Basic principles of hyperspectral imaging spectrometer
1. The working principle and structure of the system: the hyperspectral imaging spectrometer combines imaging technology and spectral technology to detect the spatial characteristics of objects while forming tens to hundreds of bands for each spatial pixel dispersion. The bandwidth is 10nm Continuous spectral coverage of about (currently the bandwidth of the SOC730 hyperspectral imaging spectrometer made in the United States has reached 2 nm). According to the scanning method of the hyperspectral imaging spectrometer, the working principle is also different. As an example of imaging by the optical imager, here is a brief description of the principle of the push-scan imaging of the focal plane detector.
1.1 The working principle of the system: the focal plane detector push-broom imaging principle, the reflected light of the ground object is imaged on the slit plane through the objective lens, and the slit acts as a light barrier to pass the image of the ground object strip in the direction of the rail, blocking other parts of the light . The radiant energy of the ground target object is collected by the objective lens and irradiated onto the dispersive element through the enhanced collimation of the slit. The dispersive element is spectrally dispersed in the direction of the vertical stripe, and the convergent lens is used to converge and image the two-dimensional CCD used in the sensor. The area array detection elements are distributed on the focal plane of the spectrometer. The horizontal direction of the focal plane is parallel to the slit, called the spatial dimension, and each row of horizontal photosensitive elements is an image of a spectral band of the ground feature; the vertical direction of the focal plane is the dispersion direction, called the spectral dimension, and each column of photosensitive elements is The ground features a spatially sampled field of view (pixel) spectral dispersion image. In this way, the image data of each frame of the area array detector is the spectral data of a feature strip in the direction of the rail, and the spectral image is continuously recorded to obtain the ground two-dimensional image.
1.2. Data acquisition system of hyperspectral imaging spectrometer: Hyperspectral imaging spectrometer is composed of optical system, signal front-end processing box and data acquisition and recording system. The playback and preprocessing of data is completed on the high-performance microcomputer through special software. The software has the following functions: data backup; fast playback; data regularization and format conversion; image segmentation and interception; standard format image data generation, etc.
Figure 2: A hyperspectral imaging spectrometer SOC710 Hyperspectral Imager with a built-in scanning device, weighing only 3Kg
Application of hyperspectral imaging spectrometer
The application range of hyperspectral imaging spectrometer covers many fields such as chemistry, physics, biology, medicine and so on. At present, hyperspectral imaging spectrometers have wide and far-reaching application prospects in land use, crop growth, classification, detection of diseases and insect pests, marine water color measurement, urban planning, petroleum exploration, core landforms and military target recognition. Visible near-infrared spectral range Hyperspectral imaging spectrometer has a broad application area for vegetation and oceans; the reflection spectrum characteristics of vegetation mainly depend on the chlorophyll content and composition in the leaves. Normally grown plants have a typical spectral shape; when poor growth, diseases and insect pests Induced lesions such as underground metal minerals will cause changes in the ratio of reflection intensity and small displacements of absorption spectrum characteristics. Observation of such displacements requires ultra-high-spectral imaging spectrometers with a spectral resolution better than 5nm and a signal-to-noise ratio of more than 100. In the light wave range, only visible light can be observed in the underwater state, and the wavelength range with good penetrability is 0.45 to 0.60 μm (blue light to yellow light), also known as the "ocean window". Visible light hyperspectral imaging spectrometer can observe the distribution of sedimentary suspended solids, plankton, chlorophyll and other sea conditions in the ocean, but to obtain information on the quality and quantity of suspended matter in the ocean surface, not only high spectral resolution, but also Very high radiation sensitivity (signal-to-noise ratio above 500).
2.1 Applications in agriculture and forestry There are many applications in agriculture and forestry, such as crop growth analysis, crop category identification, pest control analysis, yield assessment, forestry resource survey, deforestation, forest pasture survey, land desertification, soil erosion, etc.
2.1.1. Application in agriculture and forestry: the hyperspectral imaging spectrometer can be used to study the degree of influence of variety factors on wheat quality and the correlation between variety factors and quality indicators, and also to obtain the grains under environmental conditions The correlation between white matter content and wet gluten content, sedimentation value, water absorption rate, formation time and stable time, and analysis and prediction of biochemical parameters and spectral indexes of crops at different growth stages under different varieties, different fertilizer and water conditions Grain quality. On June 12, 2005, China first used a ground object spectrometer to monitor wheat stripe rust at high altitude5. In the "National Precision Agriculture Research and Demonstration Base" wheat experimental field in Xiaotangshan, Changping, the National Natural Science Foundation of China's "Surveillance and Warning of Wheat Stripe Rust Based on 3S Technology" using hot-air balloons to remotely monitor wheat stripe rust has achieved initial success This is the first time in China. This study takes wheat stripe rust as the object, according to the precise positioning of the global positioning system (GPS), uses geographic information systems (GIs) to study the prevalence of its regions, and uses remote sensing (Rs) technology to explore new real-time monitoring methods (collectively 3S technology), it is expected to eventually build a web-based wheat stripe rust monitoring and early warning information system. This achievement will provide a scientific basis for government departments to formulate wheat stripe rust control decision-making programs, and also provide new applications for the application of information technology in plant disease research. Ways to learn. The success of the project research will promote the standardization and practicalization of China's major crop disease monitoring and early warning system, realize the early detection of disease pandemics, ensure food production, increase farmers' income, and narrow the gap between China's plant disease monitoring and early warning technology and the international frontier level.
2.1.2. Agricultural crop growth monitoring: The remote sensing information of infrared band and near infrared band is mainly used, and the obtained vegetation index (NDVI) is positively correlated with the crop leaf area index and biomass. The NDVI process curve, especially the later change rate, has a good effect on predicting the yield of winter wheat, with high accuracy. In agricultural applications, high-spatial and high-spectral resolution aerial and aerospace remote sensing are used to provide timely (once every 2 to 3 days) crop growth, water and fertilizer status, and pest and disease conditions, which is called the "symptom" (Symptom) Maps) for diagnosis, decision-making, and production estimation. In order to obtain data in real time, it is necessary to repeatedly use aviation remote sensing or use various small satellites to establish a global data collection network.
The basic problems of hyperspectral remote sensing and precision agriculture research still need to be solved, such as the study of remote sensing mechanism and remote sensing signs under the action of environmental stress, the diagnosis theory of the integration of remote sensing and GIS on crop stress, and the actual distribution space of crop growth environment and harvest yield Quantitative relationship between the mechanism of difference and the effect of environmental stress and yield formation by remote sensing. In order to solve the above theoretical and application problems, it is necessary to grasp the key technologies such as hyperspectral, high resolution, radar remote sensing and other technical means and "three S" integration technology.
Analyze and estimate the biophysical parameters of vegetation such as leaf area index, biomass, total nitrogen and total phosphorus. In the research of precision agriculture, hyperspectral remote sensing has broad application prospects. For example, biophysical and biochemical parameters can be extracted from remote sensing data, which is the inversion of some important biological and agronomic parameters using high-altitude hyperspectral remote sensing data. Such research can be used to study ecosystem processes, such as photosynthesis, C, N cycle, etc., and can also be used to describe and simulate ecosystems.
The most promising and beneficial application prospect is to study the agronomic remote sensing mechanism of crop spectral characteristics and apply it to remote sensing to estimate yield, so as to achieve dynamic monitoring of crop growth, early diagnosis of diseases and insect pests and early forecast of yield. It can be used for real-time dynamic monitoring and loss assessment of agricultural natural disasters (water, drought, fire, insects, diseases, etc.), the growth of major crops, monitoring of sown area and yield forecast, grassland production estimation, grass and livestock balance estimation, and agriculture. Dynamic monitoring and evaluation of natural resources and environment, and remote sensing dynamic monitoring of cultivated land changes nationwide.
2.2. Environmental monitoring: Environmental monitoring is mainly used in 1. Petrochemical industry: such as elemental analysis in oil products, plastics, additives, catalysts, etc., and analysis and monitoring of harmful element content; 2. Ecological and environmental protection: Analysis of harmful metals in sewage or water, analysis of residual inorganic elements in plants; 3. Construction, building materials industry: combined with the identification of urban features and artificial targets, analysis of cement, glass and refractory materials.
2.3. Natural disasters and disaster assessment: At present, China is stepping up the development of environmental disaster monitoring satellites, and plans to develop a small satellite constellation composed of two optical satellites and one radar satellite by 2005. Before 2010, a small satellite constellation composed of four optical satellites and four radar satellites was developed to carry out all-weather and all-weather monitoring of the environment and disasters. Natural disaster monitoring and disaster assessment can include many types, such as floods, droughts, snow disasters, forest fires, earthquakes, and marine conditions.
Red tide refers to a phenomenon in which marine microalgae, bacteria, and protozoa overproliferate or accumulate in seawater, causing seawater to discolor. Along with economic development, coastal eutrophication has intensified. In recent years, the frequent occurrence of red tides and the continuous expansion of their scale have destroyed fishery resources and marine aquaculture. Red tide toxins have also seriously threatened human life. In 2002, a total of 79 red tides were discovered in China's waters, with a cumulative area of ​​more than 10,000 square kilometers and a direct economic loss of 23 million yuan. Using the airborne hyperspectral imaging spectrometer, 8G hyperspectral data at the scene of the red tide explosion was obtained. Through on-site sampling and post-mortem data analysis of the marine surveillance ship, the Shanghai Institute of Technology's hyperspectral imaging spectrometer uses red tide type identification software. The data quality is good and the red tide spectral characteristics are well reflected. Therefore, the data obtained by the hyperspectral imaging spectrometer can quickly respond to the red tide, which is beneficial to the early detection, classification, control and treatment of the red tide, thereby reducing the harm of the red tide.
2.4 General survey of marine resources: High-resolution imaging spectrometer can be used to obtain high-resolution images of the sea-land interaction area, which can take into account the needs of the ocean and land. At present, the hyperspectral imaging spectrometer has been applied to key coastal areas of China (Yellow River, Yangtze River) Estuary and the Pearl River Estuary) resource and vegetation surveys, coastal zone dynamic monitoring, and long-term studies of coastal zone changes. Chlorophyll distribution is an indicator closely related to ocean primary productivity, seawater eutrophication, red tide, etc. At the same time, it is also an important basis for studying global climate change. At present, the use of hyperspectral imaging spectrometer has been able to determine the chlorophyll distribution of the ocean and the distant sea more accurately, but the inversion accuracy of the chlorophyll distribution of the near-shore water body needs to be further improved.
In addition to the above practical applications, the current hyperspectral imaging spectrometer plays a major role in most fields of natural science. With the further improvement of the manufacturing technology of area array detector arrays, some new imaging spectroscopy technologies have been applied. Spectrometers with these technologies are more reliable and stable, and have small size, light weight, high spectral resolution, Better real-time performance and wider spectral range (such as the SOC710 hyperspectral imaging spectrometer made in the United States, whose spectral resolution is less than 5nm, weighs only 3kg, does not need to be equipped with a gimbal, and is very convenient for field use, see Figure 2). This hyperspectral imaging spectrometer will become the representative of a new generation of hyperspectral imaging spectrometers, and scientific researchers will also pay more attention to such spectrometers and make them more widely used.
Plastic Leveler Feet,Plastic Adjustable Feet,Furniture Leveling Feet,Adjustable Leveling Feet
Cixi Ruixin Machine Components Co., Ltd , https://www.rmlevelingfeet.com