Fixed a bug where RNL ranked filtering was not applied to images prior to creating colour maps when selected.
Batch QCPA v1.3
Fixed an error which led to images being left open after processing, subsequently causing memory overload with large datasets: https://github.com/cedricvandenberg/QCPA-batch-script
BUG ALERT: Naive Bayes clustering in QCPA
Following recent reports we can confirm that when using the Naive Bayes clustering alternative to the RNL clustering, the standard deviations instead of the means were used for downstream colour pattern analysis. This bug concerns all realeses of the toolbox
micaToolbox v2.2 release
This update mostly contains really major changes under the hood, but also a new image alignment feature: Switch to LibRAW from DCRAW, meaning it’s more future-proof and works with pretty much all the latest digital cameras. Affine align: this is
micaToolbox RAW file handling update
The micaToolbox is being updated to work with LibRAW as a replacement for DCRAW (which is no longer maintained). This means the micaToolbox will work with the latest Canon .cr3 files for example. You can download a beta release of
micaToolbox v2.1.0 release
Version 2.1.0 includes two main changes: Desaturation Function Sensor noise can create false chromatic information in dark parts of the image. We’ve therefore introduced a desaturation function available when converting an image to cone-catch. Image visualisation sRGB false colour The
BETA release of micaToolbox V2 & QCPA
The BETA version of the QCPA framework and micaToolbox version 2 is now available! This release makes a huge range of sophisticated animal vision modelling and analysis tools available to everyone. The BioRxiv pre-print manuscript detailing the QCPA framework is