【摘要】 随着机器视觉领域的研究进展和多媒体技术的迅猛发展,目标检测成了机器视觉研究方向的一个热门领域。在一张图片中,如何寻找某类特定的目标是一个十分值得研究的课题,并且具有优秀性能的算法可以用在十分广阔的领域中,比如视频中的行人、车辆检测等。文中提出了一个基于卷积神经网络的深度学习分类器和卷积的方法,该方法性能优良,漏报率较低,可以快速检测出图片中的某类目标。并且通过以行人检测为例的实验证明,甚至与传统的检测相比鲁棒性更好,速度更快,可以更好地确定标的位置。更多还原

【Abstract】 With the development of the field of computer vision and the rapid development of multimedia technology,target detection has become a hot research field of computer vision. In a picture,how to find a specific type of object is a very worthy of study. Algorithms with an excellent performance can even be used in a very wide area,such as pedestrian and vehicle detection in a vedio. This paper proposes a deep-learning classifier based on convolutional neural network( CNN) and convolution method. The experiments show that the proposed method is superior in performance and low in false negative rate. It can quickly detect some kinds of objects in the picture,and even compared with the traditional pedestrian detection,the robustness is better and the speed is faster,which can determine the position of the target better.更多还原