VIPERA 1.0 - A Versatile Imaging Platform for Education, Research and Applications Claudiu Man Electronics and Computers Transilvania University Bras ¸ov, Romˆ ania claudiu.man@student.unitbv.ro Octavian Machidon Electronics and Computers Transilvania University Bras ¸ov, Romˆ ania octavian.machidon@unitbv.ro Mihai Ivanovici Electronics and Computers Transilvania University Bras ¸ov, Romˆ ania mihai.ivanovici@unitbv.ro Abstract—We designed and implemented a versatile imag- ing platform which can be used for education, research and various applications. We used off-the-shelf components for the implementation and coded the C/C++ program required for the operation of the platform. Additional software components can be modified or added according to future needs. The imaging platform is composed of a high-speed color digital camera in Bayer format, a frame grabber and a host computer. Additionally, by using a rotating filter wheel we can use the platform for multispectral image acquisition. We created a setup with color checkers and controlled illumination for the deployment of the imaging platform in various application scenarios. We show some experimental results and draw conclusions. Index Terms—color imaging, multispectral imaging, color cor- rection, video signal acquisition I. I NTRODUCTION Image acquisition and analysis, from grayscale to color and further on to multi and hyper spectral, both in the visible spec- trum and beyond, has many technical applications in various industrial, research and educational domains. This application range has fostered the design and development of various imaging sensors and platforms in recent years that allow observing and analyzing objects of interest in a wide range of electromagnetic waves [1]. The recent expanse of the spatial data market generated research and development for imaging systems used in aerial remote sensing and mapping, using both manned and unmanned systems [2]. Cultural heritage is another field of research where multi- and hyper-spectral imaging systems have been designed and used in activities like art examinations [3], pigment identification and mapping [4] or visually enhancing old documents [5]. In agriculture, multi- and hyper-spectral imaging systems are used for vegetation and crop monitoring [6] [7] but also in the food industry for assessing various properties of food products, like meat [8] or fruits [9], or for detecting dangerous chemical compounds (e.g. in mild powders [10] or poultry [11]). Hyperspectral imaging also has medical applications, especially in tissue analysis [12] or protein tracking [13]. The general architecture of a spectral imaging platform is comprised of the imaging sensor and additional hardware for controlling the image acquisition process and storing, processing, analyzing the acquired spectral data. Given the huge amount of spectral information acquired by such systems, the performance of the underlying hardware is critical, thus reconfigurable hardware (FPGAs - Field Programmable Gate Arrays, or SoCs - Systems on Chip) or dedicated GPUs (Graphical Processing Units) are being credited as viable building blocks for multi- and hyperspectral imaging systems [14] [15]. Reconfigurable hardware offers the possibility to accelerate spectral data acquisition and, in addition, fosters computing time-consuming data preprocessing steps in the hardware (e.g. PCA - principal component analysis) - a widely-used approach for multidimensional data analysis and compression) [16]. In this paper we describe the color and multispectral imag- ing platform we designed and implemented using off-the-shelf components and custom C/C++ code. Given its versatiliy and plethora of purposes - education, research and multiple appli- cations - we call the platform VIPERA, where the acronym stands for Versatile Imaging Platform for Education, Research and Applications. In Section II we describe the platform, in Section III we show experimental results in three application scenarios and in Section IV we draw the conclusions and describe the future work. II. VIPERA 1.0 In Fig. 1 we show the block diagram of the VIPERA 1.0 system. The high-speed color digital camera is a Basler A504kc, the frame grabber is a National Instruments PCIe 1433 and the rotating filter wheel is Edmund Optics. A standard Camera Link interface is used for the communication between camera and frame grabber, PCIe interface for the transfer of the data from frame grabber to the RAM of the hosting PC, while an USB interface is used for the control of the motorized wheel. Fig. 1. VIPERA 1.0 block diagram.  Authorized licensed use limited to: UNIVERSITY OF LJUBLJANA. Downloaded on August 08,2020 at 13:38:24 UTC from IEEE Xplore. Restrictions apply.