Feature determination apparatus and method adapted to multiple object sizes

Abstract

A feature determination apparatus and method adapted to multiple object sizes are provided. The apparatus individually supplies each of the object images to a convolution neural network having severalconvolution layers to generate multiple feature maps corresponding to each object image. The apparatus calculates a feature amount of each feature image of each object image. The apparatus determinesan invalid layer start number of each object image according to a preset threshold and the feature amount corresponding to each object image. The apparatus determines a feature map extraction recommendation for each of a plurality of object sizes according to a size of each object image and the invalid layer start number of each object image.

Description

technical field [0001] The invention relates to a feature determination device and method of an object. Specifically, the present invention relates to a feature determination device and method adapting to the size of multiple objects. Background technique [0002] With the rapid development of computing hardware and deep learning technology, many computer vision devices and image processing devices (such as roadside devices) have adopted object detection models established by deep learning technology to detect objects in images or image sequences. Currently, deep learning models widely used in computer vision technology and image processing technology include convolutional neural network (Convolutional Neural Network; CNN) model, regional convolutional neural network (Region-based Convolutional Neural Network; R-CNN) model, fast Regional Convolutional Neural Network (Faster Region-based Convolutional Neural Network; Faster R-CNN) model and so on. [0003] No matter which d...

AI Technical Summary

Problems solved by technology

The memory stores multiple object images

Images