Software-Based Improvement of Image Information


taken with a CCD camera (sensor: 1600×1200 pixels). In still images extracted from these video clips smaller partial images (667×491 pixels), showing a rectangular area of 140×103 μm (=4.76 pixels per μm) containing about 50 single thrombo- cytes per image, were cropped out. Te author also received several variants of still images from Dr. M. J. Kraus, Univer- sity of Koblenz, Germany. From these partial images, squares (80×80 pixels) corresponding to 16.8×16.8 μm were cropped out. Each of these squares showed one single thrombocyte in an activated stage. In these thrombocytes the average diameter of pseudopodia was about 0.5 μm, corresponding to 2.38 pix- els. All cropped images sized to 80×80 pixels were rendered by multi-step post-processing in order to improve final image quality and fidelity of fine structures as described below. Image processing. Te following soſtware-based proce-


dures were used for successive rendering: zooming and inter- polation (Photozoom Pro, BenVista Ltd.; http://www.benvista. com/), focus stacking and deconvolution (Picolay: http:// www.picolay.de/; Combine ZP: https://combinezp.soſtware. informer.com/download/; Fitswork: https://www.fitswork.de/ soſtware/soſtw_en.php), high-dynamic range (HDR) render- ing (Photomatix Pro: https://www.hdrsoſt.com), and contour rendering (Sobel operator). Finally, additional procedures were carried out by use of standard image editing soſtware: conver- sion to a grayscale image, digital inversion, equalization of the background, and adjustment of brightness, contrast, histo- gram, and gradation. With zooming soſtware, small images can be “blown up”


Figure 2: Fibrin fibers taken from a thin-layer coverslip preparation of whole- blood. (a) Initial precipitation, (b) complex network. Images collected with an oil


immersion 100×/1.32 phase contrast objective. Image width=60 μm (a), 50 μm (b).


since the image resolution in these instruments is limited to about 0.2 μm in standard applications. Additional problems may result from the camera used dur-


ing collection of digital images if the size of the sensor, and the resulting size and number of pixels in the image are restricted. In many devices used for live cell image acquisition, video modes for slow-motion studies are especially affected because of restrictions on available sensor sizes. In images taken at the resolution limit of the optical system and/or the sensor used, several soſtware-based strategies can be helpful for enhance- ment of visual information. In this article, various methods of image reconstruction are presented and demonstrated using high-magnification images of activated thrombocytes in live cell preparations.


Material and Methods Image acquisition. Tin-layer coverslip preparations


were made from human thrombocytes gathered from throm- bocyte concentrates and examined with a dark-field micro- scope employing an oil immersion lens (100×/1.32-0.60) fitted with an iris diaphragm [5]. Time-lapse video clips were


2019 July • www.microscopy-today.com


so that fine details and contours appear with improved clar- ity and precision, especially with use of the S-spline algorithm [6,7]. A series of several images of the same feature (that is, the same specimen and equipment) taken at different focal planes are needed for focus stacking. Tese images were superimposed so that only in-focus zones were selected for reconstruction of the final image, and any out-of-focus regions were discarded. Out-of-focus regions were further eliminated by repetitive stacking using different soſtware and/or algorithms and/or by use of deconvolution soſtware, which is also based on iterative superimposition [8,9]. In over- or under-exposed zones fine details can be lost. In this case, images were duplicated for a series of different brightness/exposures so that each specimen zone was well exposed in at least one of the duplicates. Te dif- ferentially exposed images were superimposed and rendered with HDR soſtware so that a final image was generated that was free from incorrect exposure setting [10,11]. Tus, even in dark and bright zones, fine details were revealed. Te Sobel opera- tor was then used for edge detection so that fine marginal lines were accentuated [11,12]. With this combination of techniques, fine details in small pseudopodia of living thrombocytes were examined. All reconstructed images were further optimized by the common procedures mentioned above. Workflow. In principle, the workflow described can be


carried out in two ways:


Variant 1: All single-shot images selected for focus stack- ing are firstly “blown up” (that is, zoomed) by use of the S-spline algorithm. In a seconded step, the corresponding


13


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68