摘要:Shortwave infrared hyperspectral imaging technology has been widely used in various fields. Background radiation has a significant effect on the image quality and Signal to Noise Ratio(SNR).Based on the programmable shortwave infrared detector, we proposed two methods of background reference line and differential gain. The methods can be real-time monitoring of background radiation and improving the SNR of the weak signal. Experimental verification shows that the image quality and SNR of SWIR hyperspectral imaging system are better than the other systems without these methods. By using of programmable feature, frame rate of the systemis improved. High frame rate detector is necessary to high resolution hyper spectral imager.
关键词:shortwave infrared detector;background radiation;Signal to Noise Ratiol (SNR);background reference line;gain
摘要:Radiometric cross-calibration of infrared band was performed between MODIS and Tiangong-1, and then calibrated Tiangong-1 data was used to monitoring the coastal Sea Surface Temperature (SST). Quasi-simultaneous MODIS and Tiangong-1 infrared images in winter and summer acquired in 2012 and 2013, were selected for cross-calibration test. The results show that radiance of Tiangong-1 infrared band was highly correlated to that of MODIS bands in the spatially matched regions (R>0.95), and accordingly, a linear radiance-correcting function for Tiangong-1 was established against the Band 32 of MODIS. The radiometric-calibrated Tiangong-1 imageries were then used to retrieve the SST of coastal waters: the temperature were consistent with the real sea surface conditions; credited to the high spatial resolution of Tiangong-1 infrared data, the anomalies in SST due to natural or human disturbance were clearly shown, e.g., the thermal effluent of cooling waters from coastal power plant, which suggests that Tiangong-1 has the potential in water temperature monitoring for the spatially complex coastal zone.
关键词:Tiangong-1;MODIS;infrared band;cross-calibration;sea surface temperature
摘要:The first Chinese space laboratory Tiangong-1 is launched on September 29, 2011.The module carries one hyperspectral imager, which includes two imaging spectrometers: one is the visible-near infraredimaging spectrometer and the other is the short wave infrared imaging spectrometer. The hyperspectral imager has the capable of obtaining data with the nano-scale spectral resolution and high spatial resolution. For quantitative application of these data, the method and process flow of hyperspectral imager radiometric and spectral correction is proposed, based on the analysis of imaging principle and data characteristic. First, the original DN data are removed background noise, then, the data are carried out radiometric correction and bad pixels are repaired, finally, the data are performed smile correction. The scale coefficient of corrected data is also provided.
摘要:In order to obtain higher signal-to-noise ratio, hyperspectral spectrometer typically adopts motion compensation techniques to increase the integration time of detector imaging. To the mirror motion compensation mode, the imaging characteristics are analyzed and a geometric relation between mirror angle and the observation vector is established. Based on the collinearity equations, a rigorous geometric model of this kind of spectrometer can be established. The model describes the physical properties of the imaging spectrometer through the relationship between the image and the ground coordinate system. The test results indicate that the validity of the model and high-precision imaging products can be obtained.
摘要:Radiometric normalization is a key pre-processing technology for thematic mapping using hyperspectral images, which aims at eliminating response differences among the detectors. The short wave and infrared bands of the Tiangong-1 hyperspectral images were radiometric normalized using column average and standard deviation methods. Thus, three radiometric normalization algorithms including mean normalization, moment matching normalization and adjacent column balanced normalization were used for performance evaluation. Meanwhile, these three algorithms were parallel implemented on graphics processing units and compute device unified architecture. The parallel implementation methods mainly by decomposition the processing flow, which the CPU focus on procedure control and the GPU focus on data level parallel computing. A mapping model also established between the parallel computing units and the image pixels for further performance improvement. Overall, the parallel computing methods achieved a speedup about 5 times to 7 times when compared with the CPU counterparts. The proposed radiometric normalization algorithms dependent on image statistics and easy for parallel computing, which provides a thoughtful perspective on the potentials of adapting these techniques to on-board as well as on-the-ground hyperspectral image real time processing.
摘要:Tropical forest plays an important role in sustainable development, biodiversity conservation, and offers many environmental benefits, including essential ecosystem services such as climate change mitigation and adaptation. Deforestation is one of the main problems in tropical area. In this paper, we used hyperspectral data from Tiangong-1 space station platform and historical Landsat 7 ETM+ data to map forest coverage and detect forest changes in the tropical forests of Xishuangbanna, Yunnan province, China. First, pure forest pixels were extracted according to the characteristic in the histogram of red band. Then spectral similarity was calculated for each pixel with these pure forest pixels. This similarity index was used to generate forest coverage map. Then a forest change map was generated using these multi-temporal coverage maps. After evaluated with field survey data, the results showed that:(1) The hyperspectral data from Tiangong-1 can be used to map forest coverage effectively. The forest distribution pattern agreed well with ground survey data.(2) With the aid of historical forest coverage map, the forest changes can be detected. The deforestation and reforestation patches were characterized well. The recent reforestation stands in low canopy density level can also be detected.
摘要:Increasing access to means of remote sensing data, ground processing system need develop for specific types of remote sensing data. In this paper, According to Tiangong-1 sensor characteristics, high-performance parallel cluster environment based on ground-based remote sensing data preprocessing system is designed and implemented. It has quickly and efficiently handled abilities of massive remote sensing data processing tasks. The system completes the system radiometric calibration, geometric correction processing system to produce high spectral and thermal infrared spectrum of one, two remote sensing products. In the subsystem of remote sensing data product package and output, GeoTiff and HDF were chosen as the more general data format to meet many user requirements from local areas.
摘要:In this paper, the high-spatial-resolution thermal infrared domain data from the Tiangong-1 satellite were used to retrieve the Land Surface Temperature (LST) based on the radiative transfer model and Plank's law. The Heihe River Basin was selected as the study area. The retrieving process was supported by the numerous ground-based measurements from the Heihe Watershed Allied Telemetry Experimental Research (HiWATER) project implemented in the midstream of this basin. The retrieval results were available and reasonable to describe the distribution of LST in detail, which was most significant for assessing the heterogeneity of LST. To validate the retrieval results, the eight sub experimental regions with different landscapes were selected to carry out the LST ground-based observations synchronized with the Tiangong-1 satellite. The validations results showed that the retrieval LSTs fit well with ground-based LSTs, and were with a bias of -0.375℃. Additionally, the results demonstrated that the retrieval processes proposed in this paper were applicable for the Tiangong-1 satellite to retrieve the accurate LSTs.
关键词:Tiangong-1 satellite;Land Surface Temperature (LST);Validation;Heihe River Basin (HRB)
摘要:The Tiangong-1 hyperspectral imager has a thermal band with wavelength between 8.0 μm and 10.5 μm, it adopts linear array detectors and broom-push method to obtain image. In this paper, results from laboratory radiation calibration are presented. There included image adjust effects, response parameters and linearity of per detector. According to response parameters, effective quantization code and noise code, temperature range of target which can be detected and noise equivalent temperature D-value were obtained. After analyses about propagation of error, the final calibration error was calculated.
摘要:Tiangong-1 hyperspectral data has not been widely used, its data quality and application potential still need to further proof and mining practice. See5.0 data mining tool is a classification algorithm that is able to find out the hidden features of model classes in the training sample and automatically establish decision rules, Avoiding the subjectivity of artificially build the decision tree. First, by analyzing the spectral curves of different objects, select the spectral separability best band combination to Hyperspectral data dimensionality reduction,then,using See5.0 tool to generate rule sets,finally,making the classification on the same hyperspectral data from Tiangong-1 in different classification level base on rule sets and validating the results respectively by the same validation samples .In result, finding that the best separability of different feature type appeared at 655 nm and 673 nm and 781 nm and 866 nm, 984 nm of Center Wavelength respectively by spectral curve analysis,result of classification show that at level of different growth period and different subspecies of same species, classification accuracy is below 45%, showing a certain limitations, But in species classification level, classification accuracy is above 80%, showing the superiority of hyperspectral data, at the level of vegetation classification, classification accuracy can reach more than 90%.
摘要:We proposed a single-channel algorithm for land surface temperature retrieval from thermal infrared data of Tiangong-1 hyperspectral imager, the inputs of the algorithm were atmospheric water vapor content and land surface emissivity. The algorithm was evaluated by MODTRAN simulation data and ground measurements from the Heihe Watershed Allied Telemetry Experimental Research (HiWATER). The results indicated that the proposed algorithm achieved a reasonable accuracy, with a Root Mean Square Error (RMSE) of 2.72 K, which can satisfy the needs of most applications. Four Tiangong-1 thermal infrared images of Beijing city were used for analyzing the urban heat island effect, and the results indicate that Tiangong-1 thermal infrared data are suitable for the study of urban heat island effect at district level and have great potential for future applications.
关键词:Tiangong-1;land surface temperature;land surface emissivity;urban heat island effect;district scale
摘要:Hyperspectral sensing can provide richer information in earth observation, because of the technical characteristic of the combining image and spectrum. It has been applied to all aspects of the earth sciences, as a more advanced detection technology. Firstly, the author describes the imaging characteristics of hyperspectral imager, and then introduces engineering applications both in the aviation and aerospace. Finally, the author summarizes the latest development of hyperspectral imager according to the papers and research reports which published in major academic journals in domestic and foreign in recent years.
摘要:This article introduces Tiangong-1 about the design and characteristic of the hyperspectral imager, processing of orbit and control, ground data processing and applied research, and so on. The accuracy of radiation corrected and systems geometric correction products meet the needs of some application. Various applications of remote sensing verify the quality of Tiangong-1 data products. A number of valuable Achievements are got in land resources, marine, forestry, urban environmental monitoring, hydrological and other aspects of ecological monitoring. Acting as an organizer, coordinator and integrator of the Space Utilization System of CMSP, Technology and Engineering Center for Space Utilization (CSU) response organizing and managing, integration tests of hyperspectral imager, processing of orbit and control, spreading of hyperspectral data, and has been playing an import role.
关键词:Tiangong-1;Hyperspectral Imager;ground data processing;hyperspectral applications
摘要:In order to detect ship effectively, gravitational field based enhancement of ship targets was used to calculate textural and spectral index. Based on Support Vector Machine (SVM) method,the accurate relationship between space exploration information and ground objects was examined using the Tiangong-1 hyperspectral data, and used in the ships detection experiments. Results show that the ship objects can be effectively detected from the Tiangong-1 hyperspectral data against the sea water background. However, it is quite difficult to recognize the ship types from the imagery. The automatic segmentation of Tiangong-1 hyperspectral imagery is very important for ships detection. Multispectral and high resolution of the sensor is one of a main future study direction.
摘要:Based on the Tiangong-1 Thermal IR imaging and quasi synchronous Aqua MODIS data (MYD021KM), the land surface characteristic parameters were retrieved firstly. Then the surface turbulent heat fluxes and daily evapotransporation over the northern Tibetan Plateau area were derived by using Surface Energy Balance System (SEBS). The results show that the estimated surface turbulent heat fluxes and daily evapotransporation from Thermal IR imaging are in good accordance with land surface status.
摘要:Tiangong-1 hyperspectral imager is currently the highest spatial resolution and spectra composite indicator spatial spectral imager of China. To extract alteration information from Tiangong-1 hyperspectral data, an alteration information extraction method based on band correlation detection is proposed. This method was based on band correlation detection to select anomaly bands and then conducted alteration information extraction with the advantages of anomaly detection, endmember extraction, spectral matching and other methods. This new method successfully extracted the Al-OH mineral alteration anomaly information and four alteration minerals information of Tiangong-1 hyperspectral data. Findings of this research prove that the proposed method can effectively detect and extract alteration information of Tiangong-1 hyperspectral data and put forward its potential applications for geological mineral resources survey.
关键词:Tiangong-1;hyperspectral;Correlation coefficient;anomaly detection;Alteration information
摘要:Based on lucubrating the output characteristic of the infrared focal plane detector with 1500 pixels, which is made by Shanghai Institute of Technical Physics, Chinese Academy of Sciences, we develop a kind of circuit of information acquisition with the kernel of low-noise amplifier and a project scheme of data pre-processing with the kernel of FPGA. This design makes the driving circuit and the detector match well so that the high sensitivity character of the detector is fully exerted. The testing results indicates that this design plays an important role in acquiring a Noise Equivalent Temperature Difference (NETD) better than 50 mK, and the wide band quantization design guarantees the dynamic scope and abundant image levels of the system. This design is applied successfully in infrared focal plane detectors of some other kinds.
摘要:In this paper, the Tiangong-1 hyperspectral data acquired on April 18, 2013, and a Landsat TM imagery acquired on May 1, 1985 were using to monitor the wetlands changes in Yellow River estuary. By comparing the classification results of Tiangong-1 hyperspectral data and Landsat TM imagery, the result showed that the decrease in wetlands areas was largely affected by the human activities in the past 30 years, such as beach reclamation and aquaculture. Furthermore, the Tiangong-1 hyperspectral data performed better on wetlands monitoring than Landsat TM due to its higher spectral and spatial resolution.
摘要:Thermal emission from Infrared Radiation (IR) instrument is an important background source, which will reduce Signal-to-Noise Ratio (SNR) of the system. It is necessary to estimate background radiation due to instrument's thermal emission in opt-mechanical design phase. An analytic model about thermal emission of reflect/refract combined optical system is established here, which is used to calculate background of space IR systems. Object signal is compared with background to show how background affects SNR. Finally, methods about removing background are discussed.
摘要:The first imaging spectrometer developed by China, lunched onboard Tiangong-1 spacecraft, is another sensor after Hyperion of America being suitable for geological applications. It is important to evaluate the reflectance data quantitatively. However, quantitative evaluation of spaceborne data is difficult because of low spatial resolution and largegeometric error which will make it difficult to compare ground and imaging spectra. In this paper, the Tiangong-1 hyperspectral data was evaluated on the basis of HyMap data and ground spectra, and was compared with Hyperion data. Results show that, the signal-to-noise of Tiangong-1 data is larger than Hyperion data in shortwave infrared wavelength range, which is important to minerals retrieval. A correction method is put forward to reduce the spectral system error, which can improve the identification rate of minerals. Through spectral correction, the miss identification rate of Al-OH,Mg-OH/CO32- bearing minerals decrease from 71% and 67% to 29% and 28%.
摘要:Traditional hyperspectral classification methods based on per-pixel spectral or texture features fail to take account of spatial structure and spatial correlation characteristics. In order to overcome this problem, a mixed classification method is proposed which incorporates spatial information by fusing object-based segmentation with pixel-wise classifier. This paper tentatively examines two mixed classificationstrategies: (1) Combine multi-resolution segmentation algorithm which based on Fractal Net Evolution Approachwith the use of Support Vector Machine (MSVM); (2) Combine multi-scalewatershed segmentation with Support Vector Machine (WSVM). The two methods were applied to Tiangong-1 hyperspectral data and the results showed that the proposed methods improved the classification accuracy effectively which not only avoid the spectral confusion to some extent but also mitigate the land fragmentation problem.