Smart OBIA v1.0.1 - UI & Radiometric Indices Update 🚀 This patch brings critical bug fixes, UI improvements, and significantly expands the plugin's capability to process high-resolution RGB imagery (such as UAV orthomosaics and Google Earth exports) without requiring multispectral bands. 🎉 New Features & Enhancements Smart Dynamic Band Loading: The Radiometric Indices algorithm is now fully context-aware. It dynamically loads only the required bands for the selected index. You can now seamlessly run RGB-only indices on 3-band rasters without the plugin crashing or searching for non-existent NIR/SWIR bands. New RGB-Specific Indices: Added ExG (Excess Green Index) and VDVI (Visible-band Difference Vegetation Index). These are highly optimized for extracting canopies, detecting green leaves, and discriminating crops/weeds using strictly visible bands. Revamped Indices UI: The dropdown menu for index selection has been completely reorganized into logical categories: [Vegetation (RGB-Only)], [Vegetation (Requires NIR)], [Water], [Built-up], and [Moisture / Soil]. Built-in HTML Catalog: Added a dynamic Help Panel directly inside the Indices algorithm window. Users can now view a beautifully formatted HTML table containing the Description, Mathematical Formula, and Practical Application for every single supported index, without ever leaving QGIS. 🐛 Bug Fixes Classification Algorithm Crash: Resolved a critical IndentationError and syntax bug in classification_algorithm.py that prevented the module from loading in the QGIS Processing Toolbox. Missing Icons Fix: Fixed the broken UI icons. Implemented a robust, multi-level dynamic path-search system that ensures the algorithm icons (icon.png) are correctly found and rendered in the QGIS interface, regardless of the user's OS or folder structure. Index Compatibility Protection: The plugin now actively protects the user from invalid executions. If a user mistakenly selects a multispectral index (e.g., NDVI) on an RGB-only image, the algorithm will safely abort and issue a clean warning in the log instead of causing a hard Python crash.
yes
mikadishen
2026-06-10T00:10:41.924444+00:00
3.16.0
3.99.0
None
no
Plugin Tags