树莓派搭载yolov5-lite模型

Question

我按照您的讲解，并根据您提供的包安装好了onnx和opencv 但是识别单个图片提示上面错误，通过树莓派摄像头识别，卡在video窗口 识别不出东西
测试一张图片

import cv2
import numpy as np

import onnxruntime as ort
import math
import time

def plot_one_box(x, img, color=None, label=None, line_thickness=None):
    """
    description: Plots one bounding box on image img,
                 this function comes from YoLov5 project.
    param: 
        x:      a box likes [x1,y1,x2,y2]
        img:    a opencv image object
        color:  color to draw rectangle, such as (0,255,0)
        label:  str
        line_thickness: int
    return:
        no return
    """
    tl = (
        line_thickness or round(0.002 * (img.shape[0] + img.shape[1]) / 2) + 1
    )  # line/font thickness
    color = color or [random.randint(0, 255) for _ in range(3)]
    c1, c2 = (int(x[0]), int(x[1])), (int(x[2]), int(x[3]))
    cv2.rectangle(img, c1, c2, color, thickness=tl, lineType=cv2.LINE_AA)
    if label:
        tf = max(tl - 1, 1)  # font thickness
        t_size = cv2.getTextSize(label, 0, fontScale=tl / 3, thickness=tf)[0]
        c2 = c1[0] + t_size[0], c1[1] - t_size[1] - 3
        cv2.rectangle(img, c1, c2, color, -1, cv2.LINE_AA)  # filled
        cv2.putText(
            img,
            label,
            (c1[0], c1[1] - 2),
            0,
            tl / 3,
            [225, 255, 255],
            thickness=tf,
            lineType=cv2.LINE_AA,
        )

def _make_grid( nx, ny):
        xv, yv = np.meshgrid(np.arange(ny), np.arange(nx))
        return np.stack((xv, yv), 2).reshape((-1, 2)).astype(np.float32)

def cal_outputs(outs,nl,na,model_w,model_h,anchor_grid,stride):
    
    row_ind = 0
    grid = [np.zeros(1)] * nl
    for i in range(nl):
        h, w = int(model_w/ stride[i]), int(model_h / stride[i])
        length = int(na * h * w)
        if grid[i].shape[2:4] != (h, w):
            grid[i] = _make_grid(w, h)

        outs[row_ind:row_ind + length, 0:2] = (outs[row_ind:row_ind + length, 0:2] * 2. - 0.5 + np.tile(
            grid[i], (na, 1))) * int(stride[i])
        outs[row_ind:row_ind + length, 2:4] = (outs[row_ind:row_ind + length, 2:4] * 2) ** 2 * np.repeat(
            anchor_grid[i], h * w, axis=0)
        row_ind += length
    return outs



def post_process_opencv(outputs,model_h,model_w,img_h,img_w,thred_nms,thred_cond):
    conf = outputs[:,4].tolist()
    c_x = outputs[:,0]/model_w*img_w
    c_y = outputs[:,1]/model_h*img_h
    w  = outputs[:,2]/model_w*img_w
    h  = outputs[:,3]/model_h*img_h
    p_cls = outputs[:,5:]
    if len(p_cls.shape)==1:
        p_cls = np.expand_dims(p_cls,1)
    cls_id = np.argmax(p_cls,axis=1)

    p_x1 = np.expand_dims(c_x-w/2,-1)
    p_y1 = np.expand_dims(c_y-h/2,-1)
    p_x2 = np.expand_dims(c_x+w/2,-1)
    p_y2 = np.expand_dims(c_y+h/2,-1)
    areas = np.concatenate((p_x1,p_y1,p_x2,p_y2),axis=-1)
    
    areas = areas.tolist()
    ids = cv2.dnn.NMSBoxes(areas,conf,thred_cond,thred_nms) 
    return  np.array(areas)[ids],np.array(conf)[ids],cls_id[ids]

def infer_img(img0,net,model_h,model_w,nl,na,stride,anchor_grid,thred_nms=0.4,thred_cond=0.5):
    # 图像预处理
    img = cv2.resize(img0, [model_w,model_h], interpolation=cv2.INTER_AREA)
    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    img = img.astype(np.float32) / 255.0
    blob = np.expand_dims(np.transpose(img, (2, 0, 1)), axis=0)

    # 模型推理
    outs = net.run(None, {net.get_inputs()[0].name: blob})[0].squeeze(axis=0)

    # 输出坐标矫正
    outs = cal_outputs(outs,nl,na,model_w,model_h,anchor_grid,stride)

    # 检测框计算
    img_h,img_w,_ = np.shape(img0)
    boxes,confs,ids = post_process_opencv(outs,model_h,model_w,img_h,img_w,thred_nms,thred_cond)

    return  boxes,confs,ids




if __name__ == "__main__":

    # 模型加载
    model_pb_path = "num1-8.onnx"
    so = ort.SessionOptions()
    net = ort.InferenceSession(model_pb_path, so)

    # 标签字典
    dic_labels = {0: '1',
                  1: '2',
                  2: '3',
                  3: '4',
                  4: '5',
                  5: '6',
                  6: '7',
                  7: '8'}

    # 模型参数
    model_h = 320
    model_w = 320
    nl = 3
    na = 3
    stride=[8.,16.,32.]
    anchors = [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]]
    anchor_grid = np.asarray(anchors, dtype=np.float32).reshape(nl, -1, 2)
    
    # 进行推理
    img0 = cv2.imread('3.jpg')
    t1 = time.time()
    det_boxes,scores,ids = infer_img(img0,net,model_h,model_w,nl,na,stride,anchor_grid,thred_nms=0.4,thred_cond=0.5)
    t2 = time.time()
    print("%.2f"%(t2-t1))
    # 结果绘图
    for box,score,id in zip(det_boxes,scores,ids):
        label = '%s:%.2f'%(dic_labels[id],score)
        
        plot_one_box(box.astype(np.int), img0, color=(255,0,0), label=label, line_thickness=None)
    cv2.imshow('img',img0)

    cv2.waitKey(0)
    





    # img = cv2.resize(img0, [320,320], interpolation=cv2.INTER_AREA)

    # img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    # img = img.astype(np.float32) / 255.0
    # blob = np.expand_dims(np.transpose(img, (2, 0, 1)), axis=0)

    # outs = net.run(None, {net.get_inputs()[0].name: blob})[0].squeeze(axis=0)

    # nl = 3
    # na = 3
    # stride=[8.,16.,32.]
    # anchors = [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]]
    # anchor_grid = np.asarray(anchors, dtype=np.float32).reshape(nl, -1, 2)
    # model_w = 320
    # model_h = 320
    # outs = cal_outputs(outs,nl,na,model_w,model_h,anchor_grid,stride)

    # print(outs)
    # boxes,confs,ids = post_process_opencv(outs,model_h,model_w,img_h=480,img_w=640,thred_nms=0.4,thred_cond=0.5)
    # print(boxes)

测试视频

```python

import cv2
import numpy as np
import onnxruntime as ort
import time

def plot_one_box(x, img, color=None, label=None, line_thickness=None):
    """
    description: Plots one bounding box on image img,
                 this function comes from YoLov5 project.
    param: 
        x:      a box likes [x1,y1,x2,y2]
        img:    a opencv image object
        color:  color to draw rectangle, such as (0,255,0)
        label:  str
        line_thickness: int
    return:
        no return
    """
    tl = (
        line_thickness or round(0.002 * (img.shape[0] + img.shape[1]) / 2) + 1
    )  # line/font thickness
    color = color or [random.randint(0, 255) for _ in range(3)]
    c1, c2 = (int(x[0]), int(x[1])), (int(x[2]), int(x[3]))
    cv2.rectangle(img, c1, c2, color, thickness=tl, lineType=cv2.LINE_AA)
    if label:
        tf = max(tl - 1, 1)  # font thickness
        t_size = cv2.getTextSize(label, 0, fontScale=tl / 3, thickness=tf)[0]
        c2 = c1[0] + t_size[0], c1[1] - t_size[1] - 3
        cv2.rectangle(img, c1, c2, color, -1, cv2.LINE_AA)  # filled
        cv2.putText(
            img,
            label,
            (c1[0], c1[1] - 2),
            0,
            tl / 3,
            [225, 255, 255],
            thickness=tf,
            lineType=cv2.LINE_AA,
        )

def _make_grid( nx, ny):
        xv, yv = np.meshgrid(np.arange(ny), np.arange(nx))
        return np.stack((xv, yv), 2).reshape((-1, 2)).astype(np.float32)

def cal_outputs(outs,nl,na,model_w,model_h,anchor_grid,stride):
    
    row_ind = 0
    grid = [np.zeros(1)] * nl
    for i in range(nl):
        h, w = int(model_w/ stride[i]), int(model_h / stride[i])
        length = int(na * h * w)
        if grid[i].shape[2:4] != (h, w):
            grid[i] = _make_grid(w, h)

        outs[row_ind:row_ind + length, 0:2] = (outs[row_ind:row_ind + length, 0:2] * 2. - 0.5 + np.tile(
            grid[i], (na, 1))) * int(stride[i])
        outs[row_ind:row_ind + length, 2:4] = (outs[row_ind:row_ind + length, 2:4] * 2) ** 2 * np.repeat(
            anchor_grid[i], h * w, axis=0)
        row_ind += length
    return outs



def post_process_opencv(outputs,model_h,model_w,img_h,img_w,thred_nms,thred_cond):
    conf = outputs[:,4].tolist()
    c_x = outputs[:,0]/model_w*img_w
    c_y = outputs[:,1]/model_h*img_h
    w  = outputs[:,2]/model_w*img_w
    h  = outputs[:,3]/model_h*img_h
    p_cls = outputs[:,5:]
    if len(p_cls.shape)==1:
        p_cls = np.expand_dims(p_cls,1)
    cls_id = np.argmax(p_cls,axis=1)

    p_x1 = np.expand_dims(c_x-w/2,-1)
    p_y1 = np.expand_dims(c_y-h/2,-1)
    p_x2 = np.expand_dims(c_x+w/2,-1)
    p_y2 = np.expand_dims(c_y+h/2,-1)
    areas = np.concatenate((p_x1,p_y1,p_x2,p_y2),axis=-1)
    
    areas = areas.tolist()
    ids = cv2.dnn.NMSBoxes(areas,conf,thred_cond,thred_nms)
    if len(ids)>0:
        return  np.array(areas)[ids],np.array(conf)[ids],cls_id[ids]
    else:
        return [],[],[]
def infer_img(img0,net,model_h,model_w,nl,na,stride,anchor_grid,thred_nms=0.4,thred_cond=0.5):
    # 图像预处理
    img = cv2.resize(img0, [model_w,model_h], interpolation=cv2.INTER_AREA)
    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    img = img.astype(np.float32) / 255.0
    blob = np.expand_dims(np.transpose(img, (2, 0, 1)), axis=0)

    # 模型推理
    outs = net.run(None, {net.get_inputs()[0].name: blob})[0].squeeze(axis=0)

    # 输出坐标矫正
    outs = cal_outputs(outs,nl,na,model_w,model_h,anchor_grid,stride)

    # 检测框计算
    img_h,img_w,_ = np.shape(img0)
    boxes,confs,ids = post_process_opencv(outs,model_h,model_w,img_h,img_w,thred_nms,thred_cond)

    return  boxes,confs,ids




if __name__ == "__main__":

    # 模型加载
    model_pb_path = "num1-8.onnx"
    so = ort.SessionOptions()
    net = ort.InferenceSession(model_pb_path, so)
    
    # 标签字典
    dic_labels= {0:'1',
            1:'2',
            2:'3',
            3:'4',
            4:'5',
            5:'6',
            6:'7',
            7:'8'}
    
    # 模型参数
    model_h = 320
    model_w = 320
    nl = 3  #几层输出
    na = 3  #对于每一层有几种瞄框
    stride=[8.,16.,32.]
    anchors = [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]]
    anchor_grid = np.asarray(anchors, dtype=np.float32).reshape(nl, -1, 2)
    
    video = 0
    cap = cv2.VideoCapture(video)
    flag_det = False
    while True:
        success, img0 = cap.read()
        if success:
            
            if flag_det:
                t1 = time.time()
                det_boxes,scores,ids = infer_img(img0,net,model_h,model_w,nl,na,stride,anchor_grid,thred_nms=0.4,thred_cond=0.5)
                t2 = time.time()
            
                
                for box,score,id in zip(det_boxes,scores,ids):
                    label = '%s:%.2f'%(dic_labels[id],score)
            
                    plot_one_box(box.astype(np.int16), img0, color=(255,0,0), label=label, line_thickness=None)
                    
                str_FPS = "FPS: %.2f"%(1./(t2-t1))
                
                cv2.putText(img0,str_FPS,(50,50),cv2.FONT_HERSHEY_COMPLEX,1,(0,255,0),3)
                
            
            cv2.imshow("video",img0)
        key=cv2.waitKey(1) & 0xFF    
        if key == ord('q'):
        
            break
        elif key & 0xFF == ord('s'):
            flag_det = not flag_det
            print(flag_det)
            
    cap.release() 
    
    
    
    
    # # 进行推理
    # img0 = cv2.imread('3.jpg')
    # t1 = time.time()
    # det_boxes,scores,ids = infer_img(img0,net,model_h,model_w,nl,na,stride,anchor_grid,thred_nms=0.4,thred_cond=0.5)
    # t2 = time.time()
    # print("%.2f"%(t2-t1))
    # 结果绘图
    # for box,score,id in zip(det_boxes,scores,ids):
        # label = '%s:%.2f'%(dic_labels[id],score)
        
        # plot_one_box(box.astype(np.int), img0, color=(255,0,0), label=label, line_thickness=None)
    # cv2.imshow('img',img0)

    # cv2.waitKey(0)
    





    # img = cv2.resize(img0, [320,320], interpolation=cv2.INTER_AREA)

    # img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    # img = img.astype(np.float32) / 255.0
    # blob = np.expand_dims(np.transpose(img, (2, 0, 1)), axis=0)

    # outs = net.run(None, {net.get_inputs()[0].name: blob})[0].squeeze(axis=0)

    # nl = 3
    # na = 3
    # stride=[8.,16.,32.]
    # anchors = [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]]
    # anchor_grid = np.asarray(anchors, dtype=np.float32).reshape(nl, -1, 2)
    # model_w = 320
    # model_h = 320
    # outs = cal_outputs(outs,nl,na,model_w,model_h,anchor_grid,stride)

    # print(outs)
    # boxes,confs,ids = post_process_opencv(outs,model_h,model_w,img_h=480,img_w=640,thred_nms=0.4,thred_cond=0.5)
    # print(boxes)

```

Answer 1

以下内容由CHATGPT及阿里嘎多学长共同生成、有用望采纳：

---

您好！看起来您在使用树莓派搭载 YOLOv5-Lite 模型进行图像识别时遇到了一些问题。首先，让我们来解决您遇到的错误。

错误信息 ValueError: operands could not be broadcast together with shapes (3,2,40,13) (4800,2) 指出在执行 cal_outputs 函数时，两个数组的形状不兼容，无法进行元素级别的操作。具体来说，问题出现在以下这行代码：

outs[row_ind:row_ind + length, 0:2] = (outs[row_ind:row_ind + length, 0:2] * 2. - 0.5 + np.tile(grid[i], (na, 1))) * int(stride[i])

这里，np.tile(grid[i], (na, 1)) 应该将 grid[i] 复制 na 次，沿着第二个维度（索引为 1 的维度）。但是，看起来 grid[i] 的形状与 outs 数组的形状不匹配，导致无法正确地进行 tile 操作。

为了解决这个问题，您需要确保 grid[i] 的形状是 (h, w, 2)，其中 h 和 w 分别是输出特征图的高度和宽度。如果 grid[i] 的形状不正确，您需要重新生成它。以下是可能的解决方案：

1. 确保 _make_grid 函数返回正确形状的网格。

2. 检查 stride 是否正确设置，它应该能够整除 model_w 和 model_h。

3. 在 cal_outputs 函数中，确保 grid[i] 的生成是正确的，并且它的形状与输出张量的形状兼容。

另外，关于树莓派摄像头识别卡在 video 窗口的问题，这可能是由于模型推理速度跟不上视频帧率，或者是模型没有正确加载或执行。以下是一些可能的优化和调试步骤：

• 确保模型文件 num1-8.onnx 是正确的，并且与您的代码兼容。

• 检查 InferenceSession 是否正确加载了模型。

• 使用较小的输入分辨率来降低模型的计算负担。

• 确保 infer_img 函数能够在单个图像上成功运行，然后再应用到视频流上。

• 使用打印语句或日志记录来确定代码的哪一部分出现问题。

• 如果可能，尝试在树莓派上运行一些简化的代码，比如只加载模型而不进行推理，以确保基本功能正常。

由于您提供的代码较长，我无法立即提供完整的解决方案，但上述建议应该能帮助您开始调试过程。如果您需要更具体的帮助，可以提供更详细的代码片段或错误信息。

至于参考链接，以下是一些可能会有帮助的资源：

• YOLOv5 官方 GitHub - 尽管您的模型是 YOLOv5-Lite，但这里可能有通用的解决方案和讨论。
• OpenCV 文档 - 特别是如果您需要在 Android 或树莓派上使用 OpenCV。
• ONNX 运行时文档 - 用于了解如何加载和运行 ONNX 模型。

希望这些信息对您有所帮助！如果您有进一步的问题或需要更详细的指导，请随时告诉我。