ForestEyes Project: Conception, enhancements, and challenges

Rainforests play an important role in the global ecosystem. However, significant regions of them are facing deforestation and degradation due to several reasons. Diverse government private initiatives were created monitor alert for increases from remote sensing images, using different ways deal with notable amount generated data. Citizen Science projects can also be used reach same goal. consists scientific research involving nonprofessional volunteers analyzing, collecting data, their computational resources outcome advancements science increase public’s understanding problems specific knowledge areas such as astronomy, chemistry, mathematics, physics. In this sense, work presents a project called ForestEyes, which uses volunteer’s answers through analysis classification images rainforests. To evaluate quality those answers, campaigns/workflows launched Brazilian Legal Amazon results compared official groundtruth Deforestation Monitoring Project PRODES. work, first two workflows that enclose State Rondônia years 2013 2016 received more than 35,000 383 2,050 tasks only half weeks after launch. For other four workflows, even enclosing area (Rondônia) setups (e.g., image segmentation method, spatial resolution, detection target), they 51,035 volunteers’ gathered 281 3,358 tasks. performed experiments, it was possible observe achieved satisfactory overall accuracy, higher 75%, forestation non-forestation ForestEyes project. Furthermore, considering efficient better almost 66% accuracy recent deforestation, is great challenge overcome. Therefore, these show might powerful tool monitoring rainforests well obtaining high-quality labeled