Vitaly V. Krasnov

Modified temporal noise measurement method with automatic segmentation of nonuniform target, its accuracy estimation, and application to cameras of different types

Abstract. The method of measurement of imaging sensors noise by automatic segmentation of nonuniform targets is analyzed. Modifications to this method, which allows one to obtain unbiased estimation of temporal noise, calculation speed boost and separate measurement of dark and light temporal noises, are proposed. Merely two frames can be used for noise measurement with the modified method. Temporal noise measurement error of the modified method was analytically derived and experimentally verified. Light and dark temporal noises of consumer, video-surveillance, and industrial and scientific cameras were measured using the modified method.

Measurement of noises and modulation transfer function of cameras used in optical-digital correlators

Hybrid optical-digital systems based on diffractive correlator are being actively developed. To correctly estimate application capabilities of cameras of different types in optical-digital correlation systems knowledge of modulation transfer function (MTF) and light depended temporal and spatial noises is required. The method for measurement of 2D MTF is presented. The method based on random target method but instead of a random target the specially created target with flat power spectrum is used. It allows to measure MTF without averaging 1D Fourier spectra over rows or columns as is in the random target method and to achieve all values of 2D MTF instead of just two orthogonal cross-sections. The simple method for measuring the dependence of camera temporal noise on light signal value by shooting a single scene is described. Measurements results of light and dark spatial and temporal noises of cameras are presented. Procedure for obtaining cameras light spatial noise portrait (array of PRNU values for all photo sensor pixels) is presented. Results on measurements of MTF and temporal and spatial noises for consumer photo camera, machine vision camera and videosurveillance camera are presented.

Reduction of phase temporal fluctuations caused by digital voltage addressing in LC SLM "HoloEye PLUTO VIS" for holographic applications

Phase liquid crystal spatial light modulators (LC SLM) are widely used in optical applications such as real-time imaging of holograms and diffractive optical elements, which require high stability and linearity of phase modulation. However state of the art LC SLM with high resolution use digital voltage addressing scheme which, unfortunately, leads to phase fluctuations in time period of one frame. Fluctuations characteristics depend on SLM voltage addressing sequence used. We report results of measurement of phase characteristics of LC SLM “HoloEye PLUTO VIS”. This SLM is supplied with three different addressing sequences: “18-6”, “5-5” and “0-6”. Dynamics of phase fluctuations were measured for all signal values (0-255) with temporal resolution of 0.5 ms in time period of one frame for available addressing sequences. Default sequence “18-6” provided phase deviation 0.24 pi. Lowest deviation 0.07 pi was achieved with sequence “0-6”. Due to high periodicity of fluctuations it is possible to implement synchronization of SLM and registering camera or light source to reduce fluctuation effects. This was experimentally implemented using DVI video signal for synchronization of SLM and camera. With its application minimum phase deviation 0.013 pi was achieved with sequence “18-6” which is 5 times lower than achievable without synchronization.

Method of optical image coding by time integration

Method of optical image coding by time integration is proposed. Coding in proposed method is accomplished by shifting object image over photosensor area of digital camera during registration. It results in optically calculated convolution of original image with shifts trajectory. As opposed to optical coding methods based on the use of diffractive optical elements the described coding method is feasible for implementation in totally incoherent light. The method was preliminary tested by using LC monitor for image displaying and shifting. Shifting of object image is realized by displaying video consisting of frames with image to be encoded at different locations on screen of LC monitor while registering it by camera. Optical encoding and numerical decoding of test images were performed successfully. Also more practical experimental implementation of the method with use of LCOS SLM Holoeye PLUTO VIS was realized. Objects images to be encoded were formed in monochromatic spatially incoherent light. Shifting of object image over camera photosensor area was accomplished by displaying video consisting of frames with blazed gratings on LCOS SLM. Each blazed grating deflects reflecting from SLM light at different angle. Results of image optical coding and encoded images numerical restoration are presented. Obtained experimental results are compared with results of numerical modeling. Optical image coding with time integration could be used for accessible quality estimation of optical image coding using diffractive optical elements or as independent optical coding method which can be implemented in incoherent light.

Comparison of methods of suppression of undesired diffraction orders at numerical reconstruction of digital Fresnel holograms

Main information limitation of digital holograms is quantity of pixels. Necessity of spatial separation of informative diffraction order from undesired diffraction orders leads to additional reduction of quantity of resolved elements in reconstructed object. Eight methods of numerical suppression of undesired diffraction orders were compared. Under numerical testing using computer generated Fresnel holograms it was found that good quality of reconstructed images is provided by five out of eight methods. Digital Fresnel holograms were recorded and used for further methods comparison. Selection of field of frequencies zeroing method showed best results. Slightly worse results demonstrated median and Gauss filtering methods.

Generation of keys for image optical encryption in spatially incoherent light aimed at reduction of image decryption error

At present time methods of optical encryption are actively developed. The majority of existing methods of optical encryption use not only light intensity distribution, easily registered with photosensors, but also its phase distribution which require application of complex holographic schemes in conjunction with spatially coherent light. This leads to complex optical schemes and low decryption quality. To eliminate these disadvantages it is possible to implement optical encryption using spatially incoherent illumination which requires registration of light intensity distribution only. However this applies new restrictions on encryption keys: Fourier spectrum amplitude distribution of encryption key should overlap Fourier spectrum amplitude distribution of image to be encrypted otherwise loss of information is unavoidable. Therefore it seems that best key should have white spectrum. On the other hand due to fact that only light intensity distribution is registered, spectra of image to be encrypted and encryption key always have peaks at zero frequency and their heights depend on corresponding total energy. Since encrypted image contains noise, ratio of its average spectrum energy to noise average energy determines signal to noise ratio of decrypted image. Therefore ratio of amplitude at zero frequency to average spectrum amplitude (RZA) of encryption key defines decrypted images quality. For generation of encryption keys with low RZA method of direct search with random trajectory (DSRT) was used. To estimate impact of key RZA on decrypted images error numerical experiments were conducted. For experiments keys with different RZA values but with same energy value were generated and used. Numerically simulated optical encryption and decryption of set of test images was conducted. Results of experiment demonstrate that application of keys with low RZA generated by DSRT method leads to up to 20% lower error in comparison to keys generated by means of uniform random distribution.

Comparison of kinoform synthesis methods for image reconstruction in Fourier plane

Kinoform is synthesized phase diffractive optical element which allows to reconstruct image by its illumination with plane wave. Kinoforms are used in image processing systems. For tasks of kinoform synthesis iterative methods had become wide-spread because of relatively small error of resulting intensity distribution. There are articles in which two or three iterative methods are compared but they use only one or several test images. The goal of this work is to compare iterative methods by using many test images of different types. Images were reconstructed in Fourier plane from synthesized kinoforms displayed on phase-only LCOS SLM. Quality of reconstructed images and computational resources of the methods were analyzed. For kinoform synthesis four methods were implemented in programming environment: Gerchberg-Saxton algorithm (GS), Fienup algorithm (F), adaptive-additive algorithm (AA) and Gerchberg-Saxton algorithm with weight coefficients (GSW). To compare these methods 50 test images with different characteristics were used: binary and grayscale, contour and non-contour. Resolution of images varied from 64×64 to 1024×1024. Occupancy of images ranged from 0.008 to 0.89. Quantity of phase levels of synthesized kinoforms was 256 which is equal to number of phase levels of SLM LCOS HoloEye PLUTO VIS. Under numerical testing it was found that the best quality of reconstructed images provides the AA method. The GS, F and GSW methods showed worse results but roughly similar between each other. Execution time of single iteration of the analyzed methods is minimal for the GS method. The F method provides maximum execution time. Synthesized kinoforms were optically reconstructed using phase-only LCOS SLM HoloEye PLUTO VIS. Results of optical reconstruction were compared to the numerical ones. The AA method showed slightly better results than other methods especially in case of gray-scale images.

Modeling of effect of LC SLM phase fluctuations on kinoforms optical reconstruction quality

Phase-only liquid crystal (LC) spatial light modulators (SLM) are actively used in various applications. However, majority of scientific applications require stable phase modulation which might be hard to achieve with commercially available SLM due to its consumer origin. The use of digital voltage addressing scheme leads to phase temporal fluctuations, which results in lower diffraction efficiency and reconstruction quality of displayed diffractive optical elements (DOE). It is often preferable to know effect of these fluctuations on DOE reconstruction quality before SLM is implemented into experimental setup. It is especially important in case of multi-level phaseonly DOE such as kinoforms. Therefore we report results of modeling of effect of phase fluctuations of LC SLM “HoloEye PLUTO VIS” on kinoforms optical reconstruction quality. Modeling was conducted in the following way. First dependency of LC SLM phase shift on addressed signal level and time from frame start was measured for all signal values (0-255) with temporal resolution of 0.5 ms in time period of one frame. Then numerical simulation of effect of SLM phase fluctuations on kinoforms reconstruction quality was performed. Based on measured dependency, for each time delay new distorted kinoform was generated and then numerically reconstructed. Averaged reconstructed image corresponds to optically reconstructed one with registration time exceeding time period of one frame (16.7 ms), while individual images correspond to momentary optical reconstruction with registration time less than 1 ms. Quality degradation of modeled optical reconstruction of several test kinoforms was analyzed. Comparison of kinoforms optical reconstruction with SLM and numerically simulated reconstruction was conducted.

Increasing reconstruction quality of diffractive optical elements displayed with LC SLM

Phase liquid crystal (LC) spatial light modulators (SLM) are actively used in various applications. However, majority of scientific applications require stable phase modulation which might be hard to achieve with commercially available SLM due to its consumer origin. The use of digital voltage addressing scheme leads to phase temporal fluctuations, which results in lower diffraction efficiency and reconstruction quality of displayed diffractive optical elements (DOE). Due to high periodicity of fluctuations it should be possible to use knowledge of these fluctuations during DOE synthesis to minimize negative effect. We synthesized DOE using accurately measured phase fluctuations of phase LC SLM “HoloEye PLUTO VIS” to minimize its negative impact on displayed DOE reconstruction. Synthesis was conducted with versatile direct search with random trajectory (DSRT) method in the following way. Before DOE synthesis begun, two-dimensional dependency of SLM phase shift on addressed signal level and time from frame start was obtained. Then synthesis begins. First, initial phase distribution is created. Second, random trajectory of consecutive processing of all DOE elements is generated. Then iterative process begins. Each DOE element sequentially has its value changed to one that provides better value of objective criterion, e.g. lower deviation of reconstructed image from original one. If current element value provides best objective criterion value then it left unchanged. After all elements are processed, iteration repeats until stagnation is reached. It is demonstrated that application of SLM phase fluctuations knowledge in DOE synthesis with DSRT method leads to noticeable increase of DOE reconstruction quality.

Increasing quality of computer-generated kinoforms using direct search with random trajectory method

Method of increase of quality of computer generated kinoforms is proposed. It is simple direct search method similar to direct binary search method for binary holograms generation. Main difference is that proposed direct search with random trajectory method designed to process arrays with multiple phase levels. First, kinoform is generated with conventional method such as Gerchberg-Saxton. Then, elements of kinoform are sequentially switched to obtain lower normalized standard deviation (NSTD) of reconstructed image from desired image. This process goes on until minimum NSTD drop level is reached. Proposed method provides average NSTD decrease 26%.