.jpg)
A COMPUTER-IMPLEMENTED METHOD AND SYSTEM FOR DETECTING SMALL OBJECTS ON AN IMAGE USING CONVOLUTIONAL NEURAL NETWORKS
ES2908944B2
A computer-implemented method and system for detecting small objects on an image using convolutional neural networks. The method comprises: applying convolution operations (210) to an input image (102) to obtain a first set of convolutional layers (212) and an input feature map (302); analyzing the input feature map (302) to determine a first set of candidate regions (222) containing candidate objects; arranging the first set of candidate regions (222) to form a reduced feature map (228); applying convolution operations (230) to the reduced feature map (228) to obtain a second set of convolutional layers (232) and an output feature map (502); applying a Region Proposal Network (240) to the output feature map (502) to obtain a second set of candidate regions (242) containing candidate objects; classifying and applying bounding box regression (250) to each candidate region of the second set (242) to obtain, for each candidate region, a class score as a candidate object and a bounding box in the input image (102).
The present disclosure introduces a new CNN architecture for small object detection that solves the aforementioned problems, allowing the detection of small targets equal to or below 256 square pixels. This makes a big difference to the previous paper on prior art since firstly the objects of interest do not present definitive visual cues to classify them into a category and secondly the sizes of the targets considered herein disclosure are significantly smaller than those considered in prior art documents, making object detection more difficult. To detect such small objects, the overall feedforward must be low, which requires a new architecture to keep memory overhead reasonable. Furthermore, the proposed solution is an image object detector and, as such, it does not present temporal information like the video object detectors reported.