Assessing low-light cameras with photon transfer curve method

Abstract

Low-light camera is an indispensable component in various°uorescence microscopy techniques. However, choosing an appropriate low-light camera for a speci¯c technique (for example, single molecule imaging) is always time-consuming and sometimes confusing, especially after the commercialization of a new type of camera called sCMOS camera, which is now receiving heavy demands and high praise from both academic and industrial users. In this tutorial, we try to provide a guide on how to fully access the performance of low-light cameras using a well-developed method called photon transfer curve (PTC). We¯rst present a brief explanation on the key parameters for characterizing low-light cameras, then explain the experimental procedures on how to measure PTC. We also show the application of the PTC method in experimentally quantifying the performance of two representative low-light cameras. Finally, we extend the PTC method to provide o®set map, read noise map, and gain map of individual pixels inside a camera.

Characterizing and correcting camera noises in back-illuminated sCMOS cameras

“…Off i and σ i,read can be further transferred to electron unit by dividing their grey values with the camera gain value, and the latter is usually provided by camera manufactory or measured by PTC method [5]. To help compare the offset values between different cameras, we use relative offset, that is, the offset subtracted with the mean offset value of all pixels.…”

“…We calculate the precision of noise map measurement by [5]: where Map 1 and Map 2 refer to two maps measured independently under the same experimental conditions, and std is the SD of all pixels. The two maps are recommended to be measured at different days to account for experimental environment changes.…”

“…Two popular back-illuminated sCMOS cameras were tested in their best modes for low-light detection: the high gain mode for a Dhyana 95 (SN: KBS4951703002, Tucsen Photonics, China), and the 12 bit high sensitivity mode for a Prime 95B (SN: A18A203022, Photometrics, USA). The camera gains for these two modes, measured from the PTC method [5], were 1.98 DN/e-for the Dhyana 95 and 1.64 DN/e-for the Prime 95B, respectively. The maximum signal range for these two modes are ∼ 2000 e-/pixel for the Dhyana 95 and ∼ 2400 e-/pixel for the Prime 95B.…”

“…Various efforts were developed to characterize and correct the noise nonuniformity. The camera noises in sCMOS cameras (abbreviated as sCMOS noises hereafter) were characterized [5][6][7][8] and their impacts on some applications, such as single molecule localization microscopy (SMLM), were studied [9][10][11][12][13][14]. Noise correction algorithms were integrated into some commercial sCMOS cameras by their manufactories, or developed for several specific imaging scenarios by researchers [11,13,[15][16][17].…”

On-Orbit Signal-to-Noise Ratio Test Method for Night-Light Camera in Luojia 1-01 Satellite Based on Time-Sequence Imagery

“…Various on-orbit SNR test methods have been proposed so far [26,28,29,30,31], which include the variance method [32], correlation method [33], artificial neural network [34], and other less commonly used methods such as the information entropy [35] method and spatial structure method [36]. The variance method and local variance method are based on single image, the former requires large and uniform area on the image, while the latter divides less uniform image into smaller part for further processing.…”