All-round fisheye image target detection method based on cascaded neural network

Abstract

The invention relates to an all-round fisheye image target detection method based on a cascaded neural network, which is used for solving the problem of target detection difficulty caused by serious target distortion of an all-round fisheye image, and comprises the following steps: constructing a Cascade RCNN all-round fisheye image target detection network model fusing deformable convolution anddeformable pooling; training the reconstructed model; carrying out model measurement. When the network model is constructed, an improved Resnet50 network is used as a sub-network for image feature extraction, a fixed convolution and fixed pooling layer from stage 3 to stage 5 is replaced with a deformable convolution and deformable pooling layer, the receptive field and the self-learning offset can be expanded according to the state of a deformation target, effective feature extraction is carried out, and the accuracy of image feature extraction is improved. Therefore, the distortion target iseffectively modeled. Besides, the cascaded neural network is adopted, so that positive and negative samples can be subjected to multi-stage filtering for multiple times, and the accuracy of target detection is greatly improved.

Description

technical field [0001] The present invention relates to the field of computer vision, in particular to a cascaded neural network-based method for detecting a target in a look-around fisheye image Background technique [0002] Object detection is an important research content in the field of computer vision, which needs to classify and locate objects. Currently, narrow-angle pinhole cameras are mostly used for environmental perception. Their disadvantages are small field of view and blind spots. To perceive the surrounding environment, it is often necessary to use more than ten cameras, which greatly increases the processing time and is not suitable for real-time target detection processing. sexual demands. Surround-view fisheye images have a wide range of viewing angles, no blind spots, and can reduce occlusion between targets. In theory, only 4 cameras are needed to perceive 360° surrounding environment information. [0003] Although the fisheye image has a large field of...

AI Technical Summary

Problems solved by technology

[0010] Aiming at the problem that the target detection is difficult due to serious target distortion in the look-around fisheye image, the present invention provides a target detection method for look-around fish-eye images based on a cascaded neural network, which can effectively perform target detection on fish-eye images

Images