智慧农业-芝麻田杂草检测数据集】YOLO格式2类农业检测1300张高清图片数据集亮点​1300张芝麻田高清图片真实农田场景采集​txt格式标注兼容YOLO等框架​专业农学标注作物与杂草区分明确应用场景​精准除草指导智能除草机器人作业​作物监测芝麻生长状况分析数据集规格​①标注格式txt​②类别标签及中文意思0: crop —— 芝麻作物1: weed —— 杂草​③数据划分训练集909张70%验证集260张20%测试集131张10%YOLOv11 芝麻田杂草检测系统完整代码支持图片上传 / 视频检测 / 摄像头实时检测 / 一键出结果一、数据集信息标准表格1. 基础信息项目详情数据集名称芝麻田杂草检测数据集总图片数1300 张真实农田标注格式YOLO TXT 格式直接训练类别数量2 类芝麻作物 杂草应用场景智能除草、农田监测、精准农业2. 数据划分数据集数量比例训练集 Train90970%验证集 Val26020%测试集 Test13110%3. 类别信息类别ID英文标签中文标签0crop芝麻作物1weed杂草二、YOLOv11 芝麻田杂草检测 完整代码1. 数据集配置sesame.yamlpath:./sesame_datasettrain:images/trainval:images/valtest:images/testnc:2names:0:crop1:weed2. 训练代码train_sesame.pyfromultralyticsimportYOLOif__name____main__:# 加载YOLOv11模型modelYOLO(yolo11n.pt)# 训练农田小目标优化model.train(datasesame.yaml,epochs150,imgsz640,batch16,device0,lr00.01,lrf0.01,warmup_epochs3,cos_lrTrue,patience15,augmentTrue,mosaic1.0,mixup0.1,degrees15,fliplr0.5,hsv_h0.015,hsv_s0.7,hsv_v0.4,namesesame_weed_yolo11,cacheTrue,ampTrue)# 测试集评估model.val(splittest)三、一键检测系统图片 视频 摄像头3. 可视化检测工具detect_gui.py带界面支持上传图片检测上传视频检测打开摄像头实时检测一键保存结果importsysimportcv2importtorchfromultralyticsimportYOLOfromPySide6.QtWidgetsimport*fromPySide6.QtGuiimport*fromPySide6.QtCoreimport*classWorker(QThread):resultSignal(QImage)finishSignal()def__init__(self,model,source,mode):super().__init__()self.modelmodel self.sourcesource self.modemode self.runningTruedefrun(self):ifself.modeimage:imgcv2.imread(self.source)resself.model(img)frameres[0].plot()rgbcv2.cvtColor(frame,cv2.COLOR_BGR2RGB)h,w,chrgb.shape imageQImage(rgb.data,w,h,ch*w,QImage.Format_RGB888)self.result.emit(image)elifself.modevideoorself.modecamera:capcv2.VideoCapture(self.source)whileself.running:ret,framecap.read()ifnotret:breakresself.model(frame)frameres[0].plot()rgbcv2.cvtColor(frame,cv2.COLOR_BGR2RGB)h,w,chrgb.shape imageQImage(rgb.data,w,h,ch*w,QImage.Format_RGB888)self.result.emit(image)cap.release()self.finish.emit()defstop(self):self.runningFalseclassMainWindow(QMainWindow):def__init__(self):super().__init__()self.setWindowTitle(YOLOv11 芝麻田杂草检测系统)self.setFixedSize(1000,700)self.modelYOLO(best.pt)self.workerNoneself.init_ui()definit_ui(self):cQWidget()self.setCentralWidget(c)layoutQHBoxLayout(c)# 显示区域self.labelQLabel()self.label.setAlignment(Qt.AlignCenter)self.label.setStyleSheet(background-color:#1a1a1a;)layout.addWidget(self.label,stretch3)# 控制面板controlQWidget()clayoutQVBoxLayout(control)layout.addWidget(control,stretch1)self.btn_imgQPushButton(选择图片检测)self.btn_videoQPushButton(选择视频检测)self.btn_cameraQPushButton(打开摄像头)self.btn_stopQPushButton(停止)clayout.addWidget(self.btn_img)clayout.addWidget(self.btn_video)clayout.addWidget(self.btn_camera)clayout.addWidget(self.btn_stop)self.btn_img.clicked.connect(self.open_img)self.btn_video.clicked.connect(self.open_video)self.btn_camera.clicked.connect(self.open_camera)self.btn_stop.clicked.connect(self.stop_all)defopen_img(self):path,_QFileDialog.getOpenFileName()ifpath:self.workerWorker(self.model,path,image)self.worker.result.connect(self.show_img)self.worker.start()defopen_video(self):path,_QFileDialog.getOpenFileName()ifpath:self.workerWorker(self.model,path,video)self.worker.result.connect(self.show_img)self.worker.start()defopen_camera(self):self.workerWorker(self.model,0,camera)self.worker.result.connect(self.show_img)self.worker.start()defstop_all(self):ifself.worker:self.worker.stop()defshow_img(self,img):self.label.setPixmap(QPixmap.fromImage(img).scaled(self.label.size(),Qt.KeepAspectRatio))if__name____main__:appQApplication(sys.argv)winMainWindow()win.show()sys.exit(app.exec())4. 无界面快速推理predict.pyfromultralyticsimportYOLO modelYOLO(runs/detect/sesame_weed_yolo11/weights/best.pt)# 图片model.predict(test.jpg,saveTrue,conf0.25)# 视频# model.predict(test.mp4, saveTrue, conf0.25)# 摄像头# model.predict(0, showTrue, conf0.25)四、运行命令pipinstallultralytics opencv-python pyside6 torch torchvision训练python train_sesame.py启动可视化检测系统python detect_gui.py