Henry Arguello

Video anomaly recovery from compressed spectral imaging

This paper addresses the problem of video anomaly recovery from a sequence of spectrally compressed video frames. Analysis of anomalies occurring in both time and spectrum is important in video surveillance applications. We present a methodology for the recovery of anomalies such as moving objects and their spectral signatures from spectrally compressed video. The spectrally compressed video frames are obtained by using a Coded Aperture Snapshot Spectral Imaging (CASSI) system. The CASSI system encodes a 3-D data cube containing both 2-D spatial information and spectral information in a single 2-D measurement. In the proposed methodology, we use the spectrally compressed video as columns of a large data matrix G. Principal Component Pursuit (PCP) is then used to decompose G into the stationary background and a sparse matrix capturing the anomalies in the foreground. The sparse matrix is then used jointly with G to recover the spectral information of the objects of interest. An example for the recovery of video anomalies in a 3-channel spectral video system (RGB) is presented.

Snapshot Compressive Multispectral Cameras

This article presents an overview of the fundamental optical phenomena behind compressive spectral imaging systems based on coded apertures. The key mathematical concepts embodying the sensing and reconstruction methods, the framework developed to design optimal coded apertures together with the computational spectral imaging strategies for classification, unmixing, and spectral selectivity purposes are also presented. Special attention is given for describing and discussing many practical aspects of compressive spectral imaging, including discretization models of the optical system, optimal parameters design, and physical limitations. The performance of the sensing strategies and computational methods is illustrated in this article with real data and imagery for different applications. Keywords: coded aperture; compressive sensing; computational imaging; spectral imaging

Smartphone-based application for agricultural remote technical assistance and estimation of visible vegetation index to farmer in Colombia: AgroTIC

The Food and Agriculture Organization of the United Nations (FAO) has defined that agriculture is the main source of food supply given the high demand related to the population growth. Therefore, the FAO has also defined the trend to increase agricultural productivity based on the use of technology, information, and com- munication (ICT) tools. Nowadays, recent ICT tools like smartphones, cloud computing, Internet of Things (IoT), and big data support the implementation of precision agriculture to improve crops and their management. Colombia is an abundant country with suitable soils for the cultivation of diverse types of crops. However, the unknown ICT advantages for the farmers and the user rejection to adopt new technology difficult ICT implementation in the Colombian agriculture. On the other hand, the use of smartphones in various industrial applications has brought several advantages, due to their versatility, low-cost, and ease of use. Therefore, smart- phones can be used to address agriculture issues with a promissory implementation. In this sense, this work presents a smartphone application, called AgroTIC, for implementing ICT tools in the traditional work of the Colombian agricultural sector in order to support the productivity of Colombian farmers. AgroTIC takes advantage of smartphone sensors and several ICT tools. Specifically, AgroTIC is composed of four main modules: I) Communication module, II) Image processing and estimation of visible vegetation indices module, III) Production module, and IV) Marketing module. The first allows communication between farmer -to -farmer, and farmers -to -specialists/agronomist for technical assistance remote. The second module is a tool that provides an RGB analysis based on the estimation of two visible vegetation indices to help agronomist in diagnosis. The third module allows farmers to enter information related to their crops to estimate volumes of production and prepare the information for the marketing process. In the fourth module, the farmers can offer their products and establish a direct contact with different potential buyers. AgroTIC has been developed for a community of farmers who grow citrus fruits (orange, tangerine, and Tahiti lime) in the municipality of Simacota, Santander, Colombia.

Testbed Implementation of a Compressive Spectral Imaging System with Spatio Temporal Blue Noise Coded Apertures

An experimental validation of a compressive spectral imaging system using the spatio temporal blue noise (STBN) coded apertures is presented. These coded apertures are designed to better satisfy the restricted isometry property (RIP) such that improved reconstructions are attained compared to traditional random coded apertures.

Development of a Compressive Spectral Testbed based on Thin-film Color-patterned Filter Array

We report on the development of a compressive spectral imaging system based on thin-film colored patterned filter array. This array is designed and built by micrometer size thin-films with different cut-off wavelengths, and it is used as the coding element, thus permitting to attain spatial and spectral coding in a single step.

Adaptive uniform grayscale coded aperture design for high dynamic range compressive spectral imaging

Imaging spectroscopy is an important area with many applications in surveillance, agriculture and medicine. The disadvantage of conventional spectroscopy techniques is that they collect the whole datacube. In contrast, compressive spectral imaging systems capture snapshot compressive projections, which are the input of reconstruction algorithms to yield the underlying datacube. Common compressive spectral imagers use coded apertures to perform the coded projections. The coded apertures are the key elements in these imagers since they define the sensing matrix of the system. The proper design of the coded aperture entries leads to a good quality in the reconstruction. In addition, the compressive measurements are prone to saturation due to the limited dynamic range of the sensor, hence the design of coded apertures must consider saturation. The saturation errors in compressive measurements are unbounded and compressive sensing recovery algorithms only provide solutions for bounded noise or bounded with high probability. In this paper it is proposed the design of uniform adaptive grayscale coded apertures (UAGCA) to improve the dynamic range of the estimated spectral images by reducing the saturation levels. The saturation is attenuated between snapshots using an adaptive filter which updates the entries of the grayscale coded aperture based on the previous snapshots. The coded apertures are optimized in terms of transmittance and number of grayscale levels. The advantage of the proposed method is the efficient use of the dynamic range of the image sensor. Extensive simulations show improvements in the image reconstruction of the proposed method compared with grayscale coded apertures (UGCA) and adaptive block-unblock coded apertures (ABCA) in up to 10 dB.

Filtered gradient compressive sensing reconstruction algorithm for sparse and structured measurement matrices

Compressive sensing state-of-the-art proposes random Gaussian and Bernoulli as measurement matrices. Nev- ertheless, often the design of the measurement matrix is subject to physical constraints, and therefore it is frequently not possible that the matrix follows a Gaussian or Bernoulli distribution. Examples of these lim- itations are the structured and sparse matrices of the compressive X-Ray, and compressive spectral imaging systems. A standard algorithm for recovering sparse signals consists in minimizing an objective function that includes a quadratic error term combined with a sparsity-inducing regularization term. This problem can be solved using the iterative algorithms for solving linear inverse problems. This class of methods, which can be viewed as an extension of the classical gradient algorithm, is attractive due to its simplicity. However, current algorithms are slow for getting a high quality image reconstruction because they do not exploit the structured and sparsity characteristics of the compressive measurement matrices. This paper proposes the development of a gradient-based algorithm for compressive sensing reconstruction by including a filtering step that yields improved quality using less iterations. This algorithm modifies the iterative solution such that it forces to converge to a filtered version of the residual AT y, where y is the measurement vector and A is the compressive measurement matrix. We show that the algorithm including the filtering step converges faster than the unfiltered version. We design various filters that are motivated by the structure of AT y. Extensive simulation results using various sparse and structured matrices highlight the relative performance gain over the existing iterative process.

Coded Aperture Design for Compressive X-ray Tomosynthesis

This work describes an algorithm for the optimization of coded apertures in a compressive X-ray Tomosynthesis system by using the generalized mapping of the cone beam energy onto the detector. The reconstruction image quality obtained when using the optimized codes is higher compared to coded apertures generated randomly.

A random algorithm for designing the system matrix in compressive spectral imaging by homogenizing its structure

Compressive spectral imaging systems (CSI) use a focal plane array (FPA) to measure two-dimensional (2D) coded projections of a three-dimensional (3D) spatio-spectral scene. A reconstruction algorithm based on compressive sensing theory exploits the projections to retrieve the underlying 3D scene. Compressive sensing relies on two principles: sparsity and incoherence. Higher incoherence drives to better reconstructed images quality. The Colored Coded Aperture Spectral Imager (C-CASSI) is a CSI system where the coded projections are produced by optical elements named coded apertures. The C-CASSI system can be modeled as a linear transformation. The transformation matrix represents the physical effects of the coded aperture and the prism on the scene. The transformation matrix is also called the system representative matrix. The colored coded apertures modulate spatially and spectrally the light from the scene. The reconstruction image quality is highly dependent on the colored coded apertures design. An algorithm that randomly designs the coded apertures maintains the incoherence between the sensing matrix and the representation base. However, a coded aperture designed completely random, may cause the voxel information be sensed more than once, or not be sensed. This paper presents a random algorithm for colored coded apertures design by homogenizing defined parameters of the C-CASSI system representative matrix. Homogenization parameters guarantee that a voxel information would be sensed at least once. The homogenization is achieved leveling the selected parameters of the matrix, like the average of unblocking elements per column and the average of unblocking elements per row. Simulations show improvement up to 3.10 dB in the PSNR reconstructed images by using the colored coded apertures designs compared with traditional random coded apertures.

Coded aperture design for hyper-spectral image recovery via Matrix Completion

A spectral image is a 3-dimensional spatio-spectral set with a large amount of spectral information for each spatial location of a scene. Compressive Spectral Imaging techniques (CSI) permit to capture the 3D scene in 2-dimensional coded projections. The Coded Aperture Snapshot Spectral Imager (CASSI) is an optical architecture to sense a spectral image in a single projection by applying CSI. CSI increases the sensing speed and reduces the amount of collected data compared to traditional methods. The 3D scene is then recovered by solving an ℓ1-based optimization problem. However, this problem assumes that the scene is sparse in some known orthonormal basis. In contrast, a technique called Matrix Completion (MC) allows the recovery of a scene without such prior knowledge. The MC reconstruction algorithms rely on a low-rank structure of the scene. Moreover, the quality of the estimated scene from CASSI measurements depends on the coded aperture patterns used in the sensing process. Therefore, this paper proposes the design of an optimal coded aperture set for the MC methodology. The designed set is attained by maximizing the distance between the translucent elements in the coded aperture. Simulations show average improvement of around 5 dB when the designed set is used.