基于YOLO模型的垃圾组分识别影像数据集

基于YOLO模型的垃圾组分识别影像数据集元数据主要集成了城市居民在工作、生活和生产中产生的各类固体废物数据，包括纸袋、有色塑料、玻璃器皿、金属制品等29种组分，可用于训练和验证YOLO模型对不同种类垃圾组分影像的识别效果。该数据集有助于 提升相关算法城市固废图像的识别精度和效率，降低识别成本，便于固废监管部门快速、客观、准确、动态地监测城市固废堆场。

The metadata of the YOLO-based image dataset for waste component recognition primarily integrates various solid waste data generated by urban residents during work, daily life and production activities, covering 29 types of components including paper bags, colored plastics, glassware, metal products and so on. This dataset can be used to train and validate the performance of YOLO models in recognizing images of different waste components. It helps improve the recognition accuracy and efficiency of relevant algorithms for urban solid waste images, reduce recognition costs, and enable solid waste supervision departments to quickly, objectively, accurately and dynamically monitor urban solid waste stockpiles.

搜集汇总

背景与挑战

背景概述

该数据集是一个基于YOLO模型的垃圾组分识别影像数据集，整合了城市固体废物中的29种组分（如纸袋、彩色塑料、玻璃器皿和金属制品），用于训练和验证YOLO模型对垃圾图像的识别效果。数据集旨在提高城市固体废物图像识别的准确性和效率，降低识别成本，并支持固体废物监管机构快速、客观、准确地监测垃圾堆场。基于此数据集训练的YOLO模型在固体废物组分识别中准确率超过85%。