Skip to main content
Fig. 11 | International Journal of Health Geographics

Fig. 11

From: Exploring convolutional neural networks and spatial video for on-the-ground mapping in informal settlements

Fig. 11

Positive object detections after 3rd round of accuracy testing. The resolution for the image varied from 224 to 1280px. A frame window of 60 was selected for detection. Positive object detection examples for a Muddy Water (832px), b Muddy Water (1280px), c Trash (1280px), d Trash (416px), e Trash (416px), f Water Point (1280px), g Water Point (416px), h Water Point (1024px), and i Drain (416px)

Back to article page

International Journal of Health Geographics

ISSN: 1476-072X