Active Appearance Model (AAM) is an accurate and robust tool and is suitable when it’s needed to estimate shape of object when its’ approximate shape is known but varies within a certain range from instance to instance. An AAM allows complex models of shape (for example human face) and appearance to be matched to new images rapidly. An AAM contains a statistical model of the shape and gray level or color appearance of an object of interest. The associated search algorithm exploits the locally linear relationship between model parameter displacements and the residual errors between model instance and image. AAM is widely used but the research of its’ accuracy and stability still remains an important and not fully learned issue. In this paper, we study landmarks stability and error estimation produced by AAM in different lightning conditions and signal-to-noise ratio (SNR).
KEYWORDS: Probability theory, Light emitting diodes, Active optics, Error analysis, Control systems, Diodes, Computer simulations, Reliability, Buildings, Motion controllers, Image processing, Signal to noise ratio, Signal detection, Optoelectronics, Image analysis
The article is devoted to the optimization of optoelectronic systems of the spatial position of objects. Probabilistic
characteristics of the detection of an active structured mark on a random noisy background are investigated.
The developed computer model and the results of the study allow us to estimate the probabilistic characteristics of
detection of a complex structured mark on a random gradient background, and estimate the error of spatial coordinates.
The results of the study make it possible to improve the accuracy of measuring the coordinates of the object. Based on
the research recommendations are given on the choice of parameters of the optimal mark structure for use in opticalelectronic
systems for monitoring the spatial position of large-sized structures.
Using of Michelson interferometer is shown in the ï¬eld of measurement of periodical displacements of the con-trolled object. The foundations of optical interferometry are presented. The features of Michelson interferometer are described. The mathematical model of interference pattern produced by Michelson interferometer is created. It takes in consideration such parameters as the angles at which the mirrors are located and the lengths of two optical paths.
KEYWORDS: Computer programming, Detection and tracking algorithms, Digital signal processing, Image processing, Interferometers, Distortion, Optical encoders, Field programmable gate arrays, Calibration, Control systems
High precision optical encoders are used for many high end computerized numerical control machines. Main requirement for such systems are accuracy and time of measurement, therefore image processing are often performed by FPGA or DSP. This article will describe image processing algorithm for detecting and measuring pitch position on invar scale, which can be easily implemented on specified target hardware. The paper proposed to use a one-dimensional approach for pitch recognition and measure its position on the image. This algorithm is well suited for implementation on FPGA and DSP and provide accuracy 0.07 pixel.
The paper analyzes the construction matters and metrological parameters of the electrooptic converter to control linear displacements of the large structures of the buildings and facilities. The converter includes the base module, the processing module and a set of the reference marks. The base module is the main unit of the system, it includes the receiving optical system and the CMOS photodetector array that realizes the instrument coordinate system that controls the mark coordinates in the space. The methods of the frame-to-frame difference, adaptive threshold filtration, binarization and objects search by the tied areas to detect the marks against accidental contrast background is the basis of the algorithm. The entire algorithm is performed during one image reading stage and is based on the FPGA. The developed and manufactured converter experimental model was tested in laboratory conditions at the metrological bench at the distance between the base module and the mark 50±0.2 m. The static characteristic was read during the experiment of the reference mark displacement at the pitch of 5 mm in the horizontal and vertical directions for the displacement range 400 mm. The converter experimental model error not exceeding ±0.5 mm was obtained in the result of the experiment.
The main reasons of catastrophes and accidents are high level of wear of equipment and violation of the production technology. The methods of nondestructive testing are designed to find out defects timely and to prevent break down of aggregates. These methods allow determining compliance of object parameters with technical requirements without destroying it. This work will discuss dye penetrant inspection or liquid penetrant inspection (DPI or LPI) methods and computer model of microcracks image treated with fluorescent dye.
Usually cracks on image look like broken extended lines with small width (about 1 to 10 pixels) and ragged edges. The used method of inspection allows to detect microcracks with depth about 10 or more micrometers.
During the work the mathematical model of image of randomly located microcracks treated with fluorescent dye was created in MATLAB environment. Background noises and distortions introduced by the optical systems are considered in the model.
The factors that have influence on the image are listed below:
1. Background noise. Background noise is caused by the bright light from external sources and it reduces contrast on the objects edges.
2. Noises on the image sensor. Digital noise manifests itself in the form of randomly located points that are differing in their brightness and color.
3. Distortions caused by aberrations of optical system. After passing through the real optical system the homocentricity of the bundle of rays is violated or homocentricity remains but rays intersect at the point that doesn’t coincide with the point of the ideal image.
The stronger the influence of the above-listed factors, the worse the image quality and therefore the analysis of the image for control of the item finds difficulty.
The mathematical model is created using the following algorithm: at the beginning the number of cracks that will be modeled is entered from keyboard. Then the point with random position is choosing on the matrix whose size is 1024x1024 pixels (result image size). This random pixel and two adjacent points are painted with random brightness, the points, located at the edges have lower brightness than the central pixel. The width of the paintbrush is 3 pixels. Further one of the eight possible directions is chosen and the painting continues in this direction. The number of ‘steps’ is also entered at the beginning of the program. This method of cracks simulating is based on theory A.N. Galybin and A.V. Dyskin, which describe cracks propagation as random walk process. These operations are repeated as many times as many cracks it’s necessary to simulate. After that background noises and Gaussian blur (for simulating bad focusing of optical system) are applied.
In this paper, we consider the influence of various factors and interference invariant transformations measuring information on autoreflection schemes alignment control. Theoretical and experimental studies of an error for biprizm scheme. Shown that the main influencing factors are non-linear transformations in optical systems and the impact of the air path. Experimental studies were conducted based on two alignment control opto- electronic systems in which the control element (CE) is configured as one or two corner-cube retroreflectors.
Study is devoted to experimental research and development of absolute imaging position encoder based on standard
calibrated scale of invar alloy with 1 mm spacing. The encoder uses designed imaging system as a vernier and absolute
magnetic encoder as a rough indication. The features of optical design, choice and use of imaging system as long as
indexes images processing algorithm are described. A shadow method was implemented: indexes images on a CCD
array are formed by the lens focused at the scale surface; the laser module lights up the scale through a beam-splitting
prism by a parallel beam. Further dark indexes images on a light scale background are detected and analyzed to estimate the encoder position. Full range of experimental tests was set to calibrate the encoder and to estimate the accuracy. As a result, accuracy close to 1 μm at 1 m was achieved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.