Least Information Loss (LIL) Conversion of Digital Images and Lessons Learned for Scientific Image Inspection

Point Divergence Gain and Multidimensional Data Sequences Analysis

“…( a ) the dependencies of (upper) the and (lower) the vs. order of the image in the z -stack for and (middle) clustering (k-means, squared Euclidian distance, 2–6 groups) of the z -stack using connected spectra [ , ] for ; ( b ) the typical (middle) group’s images for clustering into five groups (in ( a ), middle). The original 12-bit images are visualized in 8 bits using the Least Information Loss conversion [ 23 ].…”

Žampa’s Systems Theory: A Comprehensive Theory of Measurement in Dynamic Systems

“…Therefore, we need to analyze a phenomenological discrete variable (color of an image pixel) instead of a near-continuous variable (number of photons in space). Aspects of capturing an image by a digital camera, which is a sub-system on its own, are discussed in [36]. The observed intensities represent a sum of changes of the intensities which pass a sample and a optical path of the microscope.…”

“…Vice-bit images are typically stored and visualized in a 8-bit format. As described in [36], even in a lossless compression, a series images are transformed by an algorithm which differs for each image. The alternative -a simple sectioning of original (12-or 16-bit) image into 256 levels -can leave most levels unoccupied.…”

Information Limits of Optical Microscopy: Application to Fluorescent Labelled Tissue Section

“…Instead of vectors of point information gain, vectors of point divergence gain [7] were computed from intensity histograms of the whole images for a set of α = {0.1, 0.3, 0.5, 0.7, 0.99, 1.3, 1.5, 1.7, 2.0, 2.5, 3.0, 3.5, 4.0 (the bead), 5.0, 6.0, 7.0 (the tissue)}. In case of the 2-µm bead scanned in light transmission, the z-stack which underwent the computation of the point divergence gain vectors, was firstly transformed into 8 bits by the Least Information Lost (LIL) algorithms [9] which enables to yield the maximum of information during the bit-depth reduction and to compare images through the whole stack. The clustering reduced the number of images from 258 to 81 (Imgs.…”

“…In case that the resulted image is spectrally resolved, i.e. when a colour digital camera is used, each camera channel typically detects different information [6,9,10]. The difference is generally attributed to the composition of the object which gives rise to the signal, which is ultimately true, but the exact way how this difference is conveyed by the optical path is not easy to unravel.…”