南浔区建筑垃圾遥感监测识别数据

用于实现对南浔区建筑垃圾遥感监测中问题点位的高效精准识别，主要识别内容为固体废弃物中的建筑垃圾。算法通过问题点位在影像中的相对位置，自动计算出现问题的实地坐标，便于确定需要进行实地处理的位置。并且自动识别点位问题类型，精准发现建筑垃圾与废弃建筑材料乱堆乱放等现象。为后续的人员管理与问题派发提供方便，有利于网格员及时掌握自己管理区域的情况。解决了网格员在传统人工巡查中难以发现问题，容易忽略问题，巡查范围太大，隐蔽地段、偏远地区与恶劣环境中不易巡查等问题。将系统识别出的问题点位派发给网格员，使网格员有依据、有目标的实地确认，极大的提高了发现问题的效率，节约人员时间与人工成本，避免网格员出现无效的巡查。基于无人机航拍采集的建筑垃圾遥感影像数据，通过YOLO算法进行实时目标检测。首先将单元神经网络应用于2024年9月的遥感影像，将图像分割成19x19的单元格，每个单元神经网络负责预测K个单元格。预测每个区域的概率，所有单元格上具有最大概率的类被选择并分配给特定的网格单元，生成预测点坐标（x，y），坐标系为CGCS2000，坐标为东经、北纬。
在预测类概率后，进行NMS运算，来消除不必要的锚点。算法识别下一个最高类别概率的边界框，并进行相同的运算过程，直到剩下所有不同的边界框。算法输出所需的要素，并显示各个类的边界框的细节。
抽取部分样本进行识别准确度验证，小于0.6视为识别错误，显示为FALSE；一般样本的识别准确度在0.8至1之间，大于0.6视为识别正确，显示为TRUE。通过判断结果正确或错误来纳入或排除数据，将识别正确的点位判定为建筑垃圾类别。最后将纳入的点位坐标、问题类型等信息自动上传至建筑垃圾智能监管平台，获得南浔区建筑垃圾遥感监测识别数据。

This dataset is developed for efficient and accurate identification of problematic spots in construction waste remote sensing monitoring within Nanxun District, focusing on construction waste among solid wastes. The algorithm automatically calculates the on-site field coordinates of problematic spots based on their relative positions in the images, to facilitate the confirmation of locations requiring on-site disposal. It also automatically identifies the type of each problematic spot, accurately detecting phenomena such as indiscriminate piling of construction waste and abandoned building materials. This work supports subsequent personnel management and problem assignment, enabling grid managers to timely grasp the status of their managed areas, addressing the drawbacks of traditional manual patrols including difficulty in detecting problems, easy oversight of issues, overly large patrol scope, and challenges in patrolling hidden, remote areas or harsh environments.

By assigning the problematic spots identified by the system to grid managers, the latter can conduct on-site verification with clear targets and basis, greatly improving the efficiency of problem detection, saving labor time and costs, and avoiding ineffective patrols. The dataset is constructed based on construction waste remote sensing image data collected via UAV aerial photography, with real-time object detection performed using the YOLO algorithm.

First, the remote sensing images from September 2024 are divided into 19×19 grid cells. Each grid cell is responsible for predicting K bounding boxes and the probability of each region. The class with the highest probability across all grid cells is selected and assigned to the corresponding grid cell, generating predicted point coordinates (x, y) in the China Geodetic Coordinate System 2000 (CGCS2000), with coordinates expressed as east longitude and north latitude.

After predicting the class probabilities, a Non-Maximum Suppression (NMS) operation is performed to eliminate redundant anchor boxes. The algorithm identifies the bounding box with the next highest class probability and repeats the same process until all distinct bounding boxes remain. The algorithm outputs the required elements and displays detailed information of the bounding boxes for each class.

A portion of samples are selected for recognition accuracy verification. Samples with an accuracy score lower than 0.6 are deemed recognition errors and marked as FALSE; general samples have a recognition accuracy between 0.8 and 1, while samples with an accuracy score higher than 0.6 are considered correctly recognized and marked as TRUE. Data is included or excluded based on whether the recognition result is correct or not, and correctly identified spots are classified as construction waste.

Finally, information such as the coordinates and problem types of the included spots is automatically uploaded to the construction waste intelligent supervision platform, forming the remote sensing monitoring and recognition dataset for construction waste in Nanxun District.

搜集汇总

背景与挑战

背景概述

该数据集包含503条南浔区建筑垃圾遥感监测记录，每季度更新，采用xlsx格式存储，涵盖坐标、识别准确度等字段。通过YOLO算法从无人机影像中检测建筑垃圾问题点位，提升巡查效率，应用于智能监管平台，支持精准识别和实地处理。