Image Digitalization
摘要
Analog images represent the intensity distribution of electromagnetic waves within the visible spectrum, specifically wavelengths ranging from 0.4 to 0.7 μm, commonly referred to as light. These waves can also be interpreted as a flux of energy packets named photons. In the previous chapter, we explored how to select appropriate optical components to accurately project analog images onto a plane, for measurements purposes. To convert the incoming photon flux at this plane into electrical signals, and subsequently into a two-dimensional digital output (a numerical matrix), the most commonly used devices are photodiodes. These are typically arranged in arrays, such as Charge-Coupled Devices (CCD) or Complementary Metal–Oxide–Semiconductor (CMOS) sensors array. This chapter focuses on the processes of image sampling and quantization as performed by these sensor arrays. Furthermore, spatial Fourier transform and convolution operations applied to two-dimensional signals (i.e., the image matrices) are discussed. These mathematical tools are essential for determining the optimal sensor array parameters needed to achieve high-quality digitization of analog images for measurement purposes.