Held program release Lacmus 1.0 “Furious Vaporization”. The project is a cross-platform solution for automating the search and rescue of missing people using computer vision and AI algorithms. The project is written in C# and distributed under the GPLv3 license, and its development was carried out over five years in close contact with the search and rescue community.
Software package Lacmus allows for automated search of photographs taken from a UAV, helps identify photographs that contain a lost person, and provides the user with the coordinates of the desired object. The program is based on the use of neural networks and computer vision algorithms to recognize people in images from UAVs, which makes it possible to increase the efficiency of search operations, reduce the time required to carry them out and, ultimately, increase the likelihood of successfully rescuing lost people.
Advertisement
The program supports various operating systems, including Linux, Windows and macOS, and provides the ability to use various types of graphics accelerators, such as NVIDIA, Intel, AMD, NPU and TPU GPUs, to optimize the performance of neural networks. A cross-platform framework is used for the user interface AvaloniaUIand is used to run machine learning models ONNX Runtime and lightweight version TensorFlow.
Computer vision models are distributed in the form of plug-ins, which can be downloaded from the ML models catalog in the program settings window, and each such plug-in contains the necessary dependencies, saving the end user from installing additional dependencies and libraries. The main architectures of neural networks are YOLOv5 And RetinaNet.
The project is being developed with the support of the association Lacmus Foundation, which is engaged in open research in the field of search and rescue of missing people. The organization has no legal structure and is based on a voluntary basis. Statistics show that every year thousands of people go missing around the world. Project Lacmus is intended to be the litmus test (hence its name) in solving this problem, providing search and rescue agencies with the latest tools to effectively carry out their missions and save as many lives as possible.
Thanks for reading:
Advertisement