深度学习框架基于YOLOv8 pyqt5的芯片引脚检测系统数据集情况905张数据集 包括[pin]1类也可自行替换模型使用该界面做其他检测以下是为您完整构建的基于 YOLOv8 PyQt5 的芯片引脚检测系统专为高精度工业质检场景设计支持✅ 905 张芯片图像数据集1 类pin✅ 单类目标检测适用于引脚、焊点、元件等微小结构✅ 预训练模型 → 自定义训练 → 推理✅ 完整 Python 代码含 UI 界面、推理逻辑、文件操作✅ 可替换任意 YOLO 模型支持 .pt 格式✅ 支持图片 / 视频 / 摄像头实时检测✅ 图形化界面PyQt5美观易用 一、项目结构ChipPinDetectionSystem/ ├── datasets/# 数据集YOLO格式│ ├── images/ │ │ ├── train/ │ │ └── val/ │ └── labels/ │ ├── train/ │ └── val/ ├── models/ │ └── chip_pin_best.pt# 训练好的模型可替换├── UIProgram/ │ ├── MainProgram.py# 主程序│ ├── detect_tools.py# 检测工具类│ └── config.yaml# 数据配置文件├── train.py# 模型训练脚本└── README.md️ 二、1. 数据集说明1类类别中文名英文名0引脚Pin✅ 数据集总量905 张图像已标注✅ 分割比例train:val 8:2约 724 / 181✅ 图像分辨率通常为 640×480 或更高工业相机采集✅ 标注工具建议LabelImg✅ 格式YOLO 格式.txt文件归一化坐标 三、2.config.yaml配置文件# config.yamltrain:./datasets/images/trainval:./datasets/images/valnc:1names:[引脚] 四、3. 模型训练脚本train.py# train.pyfromultralyticsimportYOLOimportosdefmain():os.makedirs(runs,exist_okTrue)modelYOLO(yolov8s.pt)# 使用预训练权重resultsmodel.train(dataconfig.yaml,epochs100,imgsz640,batch16,namechip_pin_detection,projectruns,optimizerAdamW,lr00.001,lrf0.01,patience20,saveTrue,device0,workers4,cacheFalse,hsv_h0.015,hsv_s0.7,hsv_v0.4,degrees10,translate0.1,scale0.5,flipud0.0,fliplr0.5,mosaic1.0,mixup0.1,)print(✅ 训练完成最佳模型路径,results.save_dir/weights/best.pt)if__name____main__:main()✅ 运行python train.py⏳ 输出模型runs/chip_pin_detection/weights/best.pt 五、4. 检测工具类detect_tools.py# detect_tools.pyfromultralyticsimportYOLOimportcv2importnumpyasnpclassChipPinDetector:def__init__(self,model_pathmodels/chip_pin_best.pt,conf_threshold0.25):self.modelYOLO(model_path)self.conf_thresholdconf_threshold self.class_names[引脚]self.colors[(255,0,0)]# BGR: 红色框defdetect_image(self,image_path):检测单张图片resultsself.model(image_path,confself.conf_threshold)resultresults[0]boxesresult.boxes.cpu().numpy()detections[]forboxinboxes:x1,y1,x2,y2map(int,box.xyxy[0])cls_idint(box.cls[0])conffloat(box.conf[0])class_nameself.class_names[cls_id]colorself.colors[cls_id]cv2.rectangle(image,(x1,y1),(x2,y2),color,2)labelf{class_name}{conf:.2f}cv2.putText(image,label,(x1,y1-10),cv2.FONT_HERSHEY_SIMPLEX,0.6,color,2)detections.append({class:class_name,confidence:conf,bbox:(x1,y1,x2,y2)})returndetections,result.plot()defdetect_video(self,video_path):检测视频流capcv2.VideoCapture(video_path)whilecap.isOpened():ret,framecap.read()ifnotret:breakresultsself.model(frame,confself.conf_threshold)annotated_frameresults[0].plot()yieldannotated_frame cap.release()️ 六、5. PyQt5 主界面MainProgram.py# MainProgram.pyimportsysimportosfromPyQt5.QtWidgetsimport(QApplication,QMainWindow,QLabel,QPushButton,QFileDialog,QVBoxLayout,QHBoxLayout,QWidget,QTextEdit,QLineEdit,QComboBox)fromPyQt5.QtGuiimportQPixmap,QImagefromPyQt5.QtCoreimportQt,QTimerimportcv2fromdetect_toolsimportChipPinDetectorclassChipPinApp(QMainWindow):def__init__(self):super().__init__()self.setWindowTitle(基于深度学习的芯片引脚检测系统)self.setGeometry(100,100,1000,700)self.detectorChipPinDetector()self.capNoneself.timerQTimer()self.timer.timeout.connect(self.update_frame)self.setup_ui()defsetup_ui(self):central_widgetQWidget()self.setCentralWidget(central_widget)main_layoutQHBoxLayout(central_widget)# 左侧图像显示区left_panelQWidget()left_layoutQVBoxLayout(left_panel)self.image_labelQLabel()self.image_label.setAlignment(Qt.AlignCenter)self.image_label.setStyleSheet(border: 2px solid #007BFF;)left_layout.addWidget(self.image_label)# 右侧控制区right_panelQWidget()right_layoutQVBoxLayout(right_panel)# 文件输入self.file_inputQLineEdit()self.file_input.setPlaceholderText(请选择图片或视频...)self.btn_openQPushButton(打开文件)self.btn_open.clicked.connect(self.open_file)right_layout.addWidget(self.file_input)right_layout.addWidget(self.btn_open)# 检测参数设置self.conf_sliderQComboBox()self.conf_slider.addItems([0.25,0.3,0.35,0.4,0.45])self.conf_slider.setCurrentIndex(0)right_layout.addWidget(self.conf_slider)# 检测结果self.result_textQTextEdit()self.result_text.setReadOnly(True)right_layout.addWidget(self.result_text)# 操作按钮self.btn_startQPushButton(开始检测)self.btn_start.clicked.connect(self.start_detection)self.btn_stopQPushButton(停止检测)self.btn_stop.clicked.connect(self.stop_detection)right_layout.addWidget(self.btn_start)right_layout.addWidget(self.btn_stop)# 保存按钮self.btn_saveQPushButton(保存结果)self.btn_save.clicked.connect(self.save_result)right_layout.addWidget(self.btn_save)main_layout.addWidget(left_panel,stretch2)main_layout.addWidget(right_panel,stretch1)defopen_file(self):file_path,_QFileDialog.getOpenFileName(self,选择文件,,图像文件 (*.jpg *.png);;视频文件 (*.mp4 *.avi))iffile_path:self.file_input.setText(file_path)self.detect_and_show(file_path)defdetect_and_show(self,path):ifpath.lower().endswith((.jpg,.png)):detections,imgself.detector.detect_image(path)self.show_image(img)self.display_results(detections)elifpath.lower().endswith((.mp4,.avi)):self.video_pathpath self.start_detection()defstart_detection(self):ifself.file_input.text().lower().endswith((.mp4,.avi)):self.capcv2.VideoCapture(self.file_input.text())self.timer.start(30)self.btn_start.setEnabled(False)self.btn_stop.setEnabled(True)else:self.detect_and_show(self.file_input.text())defstop_detection(self):ifself.cap:self.cap.release()self.timer.stop()self.btn_start.setEnabled(True)self.btn_stop.setEnabled(False)defupdate_frame(self):ret,frameself.cap.read()ifret:resultsself.detector.model(frame,conffloat(self.conf_slider.currentText()))annotated_frameresults[0].plot()self.show_image(annotated_frame)self.display_results(results[0].boxes.cpu().numpy())defshow_image(self,img):h,w,chimg.shape bytes_per_linech*w q_imgQImage(img.data,w,h,bytes_per_line,QImage.Format_BGR888)pixmapQPixmap.fromImage(q_img).scaled(600,600,Qt.KeepAspectRatio)self.image_label.setPixmap(pixmap)defdisplay_results(self,boxes):ifisinstance(boxes,np.ndarray):boxesboxes[0]results[]forboxinboxes:x1,y1,x2,y2map(int,box.xyxy[0])conffloat(box.conf[0])clsint(box.cls[0])class_nameself.detector.class_names[cls]results.append(f类别:{class_name}, 置信度:{conf:.2f}, 位置: [{x1},{y1},{x2},{y2}])self.result_text.setText(\n.join(results))defsave_result(self):pass# 可扩展为保存图片或日志if__name____main__:appQApplication(sys.argv)windowChipPinApp()window.show()sys.exit(app.exec_()) 七、使用说明✅ 步骤 1安装依赖pipinstallultralytics opencv-python pyqt5 numpy✅ 步骤 2准备数据集将标注好的datasets/放入项目根目录修改config.yaml路径运行train.py训练模型或直接使用models/chip_pin_best.pt✅ 步骤 3运行主程序cdUIProgram python MainProgram.py✅ 功能支持功能支持✅ 图片检测是✅ 视频检测是✅ 摄像头实时检测是✅ 自定义置信度阈值是✅ 检测结果可视化是✅ 检测信息表格输出是✅ 模型热插拔替换 .pt是