China
Africa
Explore chapters and articles related to this topic
Regression Analysis Methods for Agri-Food Quality and Safety Evaluations Using Near-Infrared (NIR) Hyperspectral Imaging
Published in Surajbhan Sevda, Anoop Singh, Mathematical and Statistical Applications in Food Engineering, 2020
Chandra B Singh, Digvir S Jayas
Hyperspectral imaging is an imaging technique that collects the full spectra of each pixel of the scanned sample. This spatially distributed spectral information of scanned material, termed as a hypercube, is described by a three-dimensional matrix of size m × n × λ, where m and n are the spatial dimensions (pixels) in x and y coordinates and λ is the spectral dimension (wavelength) (Fig. 3). Dimensionality of hypercube is reduced by applying image processing (in spatial dimension) combined with analytical tools in chemometrics (in spectral dimension) and distinct featural information is extracted in order to develop classification and calibration models (Grahn and Geladi, 2007; Geladi and Grahn, 1996). Multivariate image analysis (MVI), which uses PCA as the data reduction and feature extraction tool, is popularly used in hyperspectral imaging. The hyperspectral imaging data (hypercube) is reshaped into a two-dimensional array by sequentially rearranging all the spatial information (intensities) at each of the λ wavelengths into a column, as shown in Fig. 4 for scanned grain kernels. This results in a (m × n) × λ sized two-dimensional array, where m × n is the total number of pixels in a sample. The PCA, as explained in the previous section, is then applied to the reshaped two-dimensional data and each pixel (row) becomes a sample and wavelength (column) becomes a variable. The PCA transforms this matrix into eigenvalues, eigenvectors and PC scores which are selectively used for calibration and classification.
Advanced Light Microscopy Techniques
Published in Jay L. Nadeau, Introduction to Experimental Biophysics, 2017
In widefield fluorescence microscopy, an AOTF may be used as an emission filter, scanning across the visible range in selected increments from 2 to 50 nm and obtaining an image at each wavelength. This technique is called hyperspectral imaging because each image is a four-dimensional “cube,” where the “axes” are x, y, z, and λ. The advantages of this technique over RGB imaging can readily be imagined. It provides a full spectral signature for every pixel in the image, allowing for easy unmixing of multiple fluorophores, even those with highly overlapping spectra (Figure 8.1). The disadvantages of hyperspectral imaging are large data set size and very slow acquisition. Depending upon the brightness of a sample and the number of wavelengths desired, collecting and saving a single cube can take up to several minutes. Implementing this technique is as simple as purchasing an AOTF and installing it just before the CCD in the optical path. The only advantage to purchasing a complete commercial system is the software, though custom software often performs better if the skills are available to write it.
Application of Image Processing and Data Science in Advancing Agricultural Design
Published in Ankur Dumka, Alaknanda Ashok, Parag Verma, Poonam Verma, Advanced Digital Image Processing and Its Applications in Big Data, 2020
Ankur Dumka, Alaknanda Ashok, Parag Verma, Poonam Verma
Hyperspectral imaging is a spectral imaging technique which employs each pixel of image for acquiring set of images with certain spectral bands. This technique combines the advantages of optical spectroscopy as an analytical tool with two-dimensional (2-D) object visualization obtained by optical imaging. Thus, the sensor samples the hyperspectral cube in four different ways as spatial scanning, snapshot imaging, spectral scanning, and spatio-spectral scanning.
Hyperspectral anomaly detection: a performance comparison of existing techniques
Published in International Journal of Digital Earth, 2022
Noman Raza Shah, Abdur Rahman M. Maud, Farrukh Aziz Bhatti, Muhammad Khizer Ali, Khurram Khurshid, Moazam Maqsood, Muhammad Amin
In recent years, hyperspectral imaging has gained a lot of attention because of its spectral and spatial resolution (Goetz et al. 1985). Its spectral information ranges from visual band to far-infrared spectrum and provides a spectral curve of materials (Fauvel et al. 2013; Tao et al. 2019). In HSI, each image is a multi-spectral cube that has two spatial dimensions and one spectral dimension. The spectral dimension contains a high-dimensional reflectance vector (Li et al. 2018; Xie et al. 2019a). Depending on the HSI sensor, the hyperspectral data cube can be composed of several hundred spectral bands, each band with a narrow range of wavelengths, i.e. 10–20 nm. Due to the high information content in its spectral bands, HSI is used in applications, such as remote sensing (Khan et al. 2018). HSI can also distinguish between different materials based on their spectral signatures (Li et al. 2018). This property is exploited in various applications (Xie et al. 2019a) such as image classification (Camps-Valls and Bruzzone 2005; Harsanyi and Chang 1994), minerals detection (Reed and Yu 1990), tracking changes in the environment (Bioucas-Dias et al. 2012; Theiler and Wohlberg 2012) and anomaly detection (Reed and Yu 1990; Stein et al. 2002) etc.
A fresh look at computer vision for industrial quality control
Published in Quality Engineering, 2022
Bart De Ketelaere, Niels Wouters, Ioannis Kalfas, Remi Van Belleghem, Wouter Saeys
In order to broaden the scope of computer vision, the interest in systems that allow to “see” more than the human eye has spurred over the last two decennia (Lu et al. 2020). One of the most promising directions is the development of multi- and hyperspectral imaging systems. Their main idea is to build imaging systems that expand the working principle of RGB cameras in two directions.
Inexpensive multispectral imaging device
Published in Instrumentation Science & Technology, 2022
Multi/hyperspectral imaging is an advanced and precise remote sensing technique that includes two or more bands in the visible and invisible spectrum.[1,2] It is an important imaging technique to obtain color data and spectral monitoring.[3,4] Using more than three primary colors provides an expanded color gamut and minimization of the observation errors.[5] While using an invisible spectrum for processing an image, it is possible to obtain more data by increasing the band number of the image.[6,7] These imaging devices are widely used in space observation applications and the defense industry.[8,9] Recently these devices are becoming more popular in civil applications as well.[10,11] The popularity of these devices is increasing for image acquisition and processing applications conducted in industrial processes and research laboratories.[5,6] For example, line scanners and area imaging sensors with visible and near-infrared bands are frequently used options for evaluating and predicting the shape, area, weight, and defect properties of food products.[12,13] Another application for line scanners is to characterize metals with the help of the reflectance of the materials in specific wavelengths. In health practice, these devices are used for non-contact skin disease detection.[14,15] In agriculture, vegetation diseases, water stresses and nitrogen balance are determined based on red edge and near infrared spectral bands. Thus, the health motoring, growth rate estimation, and hydration status of the plants are evaluated with more accuracy compared to visible color observations.[11,16,17] These studies claim that increasing the observed wavelength is offering detailed information on an inspected object.[18–20] On the other hand, the application of advanced technologies in food science, agriculture, life sciences, and mineral sciences is a need of high-quality research.[21,22]