Introduction To Modern Photogrammetry

Recently we have concluded that image-based features derived from the microenvironment have an enormous impact on successfully determining the class of an object1. Here we demonstrate that deep learning-based phenotypic analysis of cells with a properly chosen microenvironment-size provides results comparable to our earlier neighbourhood-based methods that utilise hand-crafted image features. We hypothesised that treating cells with equal weight, regardless of their position within the cellular microenvironment, is suboptimal, and direct neighbours have a larger impact on the phenotype of the cell-of-interest than cells in its larger proximity. Hence we present a novel approach that (1) considers the fully featured view of the cell-of-interest, (2) includes the neighbourhood and (3) gives lesser weight to cells that are far from the cell. To achieve this, we present a transformation similar to those characteristic for fisheye cameras. Such a transformation satisfies all the above defined criteria, with a fast rate of transform for any images. Using the proposed transformation with proper settings we could significantly increase the accuracy of single-cell phenotyping, both in case of cell culture and tissue-based microscopy images. The range of potential applications of the proposed method goes beyond microscopy, as we present improved results on the iWildCam 2020 dataset containing images of wild animals.

Show me your neighbour and I tell what you are: fisheye transformation for deep learning-based single-cell phenotyping

Incorporating information about the surroundings can have a significant impact on successfully determining the class of an object. This is of particular interest when determining the phenotypes of cells, for example, in the context of high-throughput screens. We hypothesized that an ideal approach would consider the fully featured view of the cell of interest, include its neighboring microenvironment, and give lesser weight to cells that are far from the cell of interest. To satisfy these criteria, we present an approach with a transformation similar to those characteristic of fisheye cameras. Using this transformation with proper settings, we could significantly increase the accuracy of single-cell phenotyping, both in the case of cell culture and tissue-based microscopy images, and we present improved results on a dataset containing images of wild animals. 

https://doi.org/10.1016/j.crmeth.2022.100339

Fisheye transformation enhances deep-learning-based single-cell phenotyping by including cellular microenvironment

Nowadays, mobile phones are more than a device that can only satisfy the communication need between people. Since fisheye lenses integrated with mobile phones are lightweight and easy to use, they are advantageous. In addition to this advantage, it is experimented whether fisheye lens and mobile phone combination can be used in a photogrammetric way, and if so, what will be the result. Fisheye lens equipment used with mobile phones was tested in this study. For this, standard calibration of ‘Olloclip 3 in one’ fisheye lens used with iPhone 4S mobile phone and ‘Nikon FC-E9’ fisheye lens used with Nikon Coolpix8700 are compared based on equidistant model. This experimental study shows that Olloclip 3 in one fisheye lens developed for mobile phones has at least the similar characteristics with classic fisheye lenses. The dimensions of fisheye lenses used with smart phones are getting smaller and the prices are reducing. Moreover, as verified in this study, the accuracy of fisheye lenses used in smartphones is better than conventional fisheye lenses. The use of smartphones with fisheye lenses will give the possibility of practical applications to ordinary users in the near future.

/pdf/the-geometry-and-usage-of-the-supplementary-fisheye-lenses-crsbg2cyko.pdf

Cumhur Sahin

Papers

The Geometry and Usage of the Supplementary Fisheye Lenses in Smartphones