China
Africa
Explore chapters and articles related to this topic
Optical Sensing
Published in Araz Yacoubian, Optics Essentials, 2018
One example of imaging sensors is the imaging spectrometer, where the spectrum of an image is obtained using an imaging apparatus that is coupled to a spectral measurement apparatus. These techniques are called hyperspectral imaging, where a three-dimensional data is obtained, two-dimensional for image in the x-y space, and a third dimension would be wavelength. A coarse variation of this is multispectral imaging, where the wavelength bands are not as narrow as a hyperspectral imaging, and the images are taken either by multiple camera systems or using a single camera in combination with optical filters.
Remote sensing data quality model: from data sources to lifecycle phases
Published in International Journal of Image and Data Fusion, 2019
Árpad Barsi, Zsófia Kugler, Attila Juhász, György Szabó, Carlo Batini, Hussein Abdulmuttalib, Guoman Huang, Huanfeng Shen
Optical sensors are the most common RS instruments and photographic cameras are the oldest sensors in EO. In the first era of EM radiation detection, film was placed in the focal plane behind the optics mounted in a camera to obtain images. In the last decades, digital sensors (CCD/CMOS) have widely replaced this technology (Zhao 2003). Obtained panchromatic and monochrome images were certainly a perfect solution for geometric data acquisition in mapping, but the need for spectral information to classification processes was an essential step forward. Filters, beam splitting and multi-camera heads were the major technological solutions to obtain multispectral images. Radiometer is an instrument designed to measure the intensity of electromagnetic radiation in a set of wavebands reaching from visible light to infrared radiation. The electro-optic radiometers are similar in design to cameras, but instead of film, they use an electronic detector to record the intensity of electromagnetic energy. Radiometers that measure more than one waveband are called multispectral radiometers. Light is separated into discrete wavebands to obtain multiple waveband or multichannel data. This separation can be carried out using filters, prisms or other sophisticated techniques (Srivastava et al. 2014). Hyperspectral sensors or imaging spectrometers are instruments that acquire images in many narrow contiguous spectral bands throughout the visible, and infrared portion of the spectrum (Qian et al. 2004). They collect data in up to hundreds of bands enabling the construction of continuous reflectance measures (Lillesand et al. 2015).
Hybrid optimal joint spatial-spectral hyperspectral image classification using modified DHO-based GIF with JRKNN
Published in The Imaging Science Journal, 2023
M. Krishna Satya Varma, K. Raja, N. K. Kameswara Rao
In the current era, it has become relatively simple to obtain HSI and remote sensing images with great spectral and spatial resolution. HSIs have a wide range of applications in the medical domain [1], mining [2], military [3], and environmental domain [4], due to their high-resolution capability for fine spectra. The collection of HSIs is purely dependent on imaging spectrometers placed at various locations. In the early 1980s, the imaging spectrum was created. It is utilized to capture electro-magnetic waves in the mid-infrared, near-infrared, visible, and ultraviolet-based image areas. Here, the imaging spectrometer can capture numerous continuous and extremely narrow band images, where each pixel of the image is developed by a standard wavelength range and each pixel is formed with a completely emitted or reflected spectrum. As a result, HSIs have a lot of information, many bands, and excellent spectral resolution. The most used hyperspectral RSI processing techniques include dimensionality reduction and classification [5], transformation [6], noise reduction [7], and image correction [8]. The hyperspectral field's most active component of research is HSI classification [9] for better outcomes. Computer-aided classification of HSIs includes the environment, identifying and extracting features, as well as identifying and classifying information about the earth's surface. The pixels are classified based on their spectral properties into one of many classes through the HSI classification. Furthermore, the data acquired by HSIs becomes more comprehensive and detailed as hyperspectral imaging systems become more advanced, and these systems also enhance spatial and spectral resolution. Thus, the data, size, and bands of HSIs are varied based on application specifics, as some applications require low resolution, and some applications require higher resolution. The number of spectral imaging bands displayed in HSIs is greater than that in standard multispectral images, and the HSI have a greater capacity to analyse objects due to their improved spectral resolution.