ALaM Project
The DOST-Advanced Science and Technology Institute-Automated Labeling Machine (ASTI-ALaM) Project is the DOST-ASTI component of the Philippine Sky Artificial Intelligence Program (SkAI-Pinas). The project aims to develop an optimized workflow in developing machine learning and artificial neural network-based models for different application domains, as well as to develop an online model store where the developed models of the project can be accessed for tasks such as mapping and computer vision using traditional remote sensing, machine learning, and deep learning.
The ASTI-ALaM Project’s primary objective is to maximize the utilization of the country's available remote sensing data, taking advantage of the huge volume of archived datasets that is already available in both raw and processed format to serve as base sets of training data which can be further augmented to train more accurate deep learning models.
The project involves (1) segmenting objects within satellite-captured RGB images, classifying them into natural or man-made features, and (2) estimating the structural strengths of buildings using satellite-captured images.
The current work of the ASTI-ALaM team includes (1) the full-scale training and optimization of AI models, (2) act as the interagency interface with UP Mindanao’s ALaM-Large-Scale Initiative (ALaM-LSI), and (3) support ALaM-LSI in AI design and Artificial Intelligence Programming Interface (AIPI) framework implementation.
By developing AI-based solutions that can be applied in different fields, such as road management, urban planning, and disaster management, the project aims to provide decision-makers with invaluable tools to enhance their decision-making process using AI.
CONTACT US
ALaM Project