基于深度学习的智能分类垃圾桶 Intelligent classification trash can based on deep learning I中文摘要随着人类科技的进步和生活质量的提高,随之而产生的生活垃圾也越来越多,因此如何有效的回收处理生活垃圾成为人们关注的焦点。调查研究发现,在源头处对垃圾进行分类处理的方法是整个垃圾分类处理流程中最高效的也是分类最彻底的。而当前的分类规则其一是难以做到各地全部统一,另一方面是垃圾分类种类繁多,难以区分,会对人的生活质量造成一定影响。目前市面上出现了种类繁多的“智能垃圾桶”,但基本都只是实现了自动开合功能,不具备垃圾分类识别的能力。而当前的人工智能技术正在飞速发展,深度学习领域的图像识别方向有了长足的发展,使得使用图像识别技术对垃圾进行分类成为可能。选择正确的图像识别分类算法是本项目的重中之重。因此,本文对过往的图像识别方面的突出贡献的算法做介绍和总结,把握其发展的脉络,从而说明选择 MobileNetV2 网络的原因。在本项目中,系统可分为垃圾识别模块和垃圾分类投放模块,即识别模块和控制模块。在垃圾投入到垃圾桶中,由识别模块对垃圾进行识别分类,再把分类结果发送给控制模块,由控制模块将其投放到对应的垃圾桶中。经过实际测试,垃圾的类别判断正确率能在可接受范围内,而识别分类速度则在经过分类结果滤波后,平均成功识别一次垃圾的平均时间为 4 秒。这证明该项目在智能垃圾分类领域还是很有发展前景的。关键词:垃圾分类,卷积神经网,MobileNetVIIAbstractWith the advancement of human science and technology and the improvement of quality of life, more and more domestic waste is generated. Therefore, how to effectively recycle and treat domestic waste has become the focus of people's attention. The investigation and study found that the method of sorting waste at the source is the most efficient and the most thorough in the entire waste sorting process.One of the current classification rules is that it is difficult to unify all regions, on the other hand, there are many types of garbage classification, which are difficult to distinguish, which will have a certain impact on people's quality of life....