Abstract: The purpose of object detection is to recognize the category of specific objects and the position in the image, which can be applied to the field of artificial intelligence such as pedestrian, object, face detection and intelligent transporta -tion. Nowadays, neural network technology based on deep learning has been gradually integrated into people's life, and deep convolution neural network has also made remarkable achievements in various fields of computer vision. This paper will introduce the research progress of target detection algorithm based on deep learning, the characteristics of common data sets and the key parameters of performance evaluation. It will systematically describe the main target detection algorithms, the basic framework, advantages and disadvantages, and prospect the problems and research trends of current target detection algorithms.

Key words: Computer vision; Object detection; Deep learning; Causal reasoning

摘要： 目标检测的目的是识别图像中特定物体的类别和图像中的位置，实际可以应用于行人、物体、人脸检测和 智能交通等人工智能领域，现如今，基于深度学习的神经网络技术已经逐渐融入人们的生活，深度卷积神经网络也 在计算机视觉各个领域取得了显著的成就。文章将介绍基于深度学习的目标检测算法的研究进展、常用数据集特点 以及性能指标评价关键参数，系统阐述目标检测主流算法的提出、基本框架以及优缺点，并对当前目标检测算法的 问题和研究趋势进行了展望。

关键词: 计算机视觉；目标检测；深度学习；因果推理