如何在GoCV中提取信心

Question

我尝试用Golang重新制作Python ver Mask检测器，但我在使用GoCV时遇到了“提取置信度”的问题，python代码是这样说的

detections = net.forward()

# loop over the detections
for i in range(0, detections.shape[2]):
    # extract the confidence (i.e., probability) associated with
    # the detection
    confidence = detections[0, 0, i, 2]

它在python上工作得很好，但是我怎么才能用GoCV获得置信度呢？

这是我目前的工作

    currPath := "/media/fz/Project (Running)/Unknown/TryDetectMask/face_detector"
    prototxtPath := path.Join(currPath, "deploy.prototxt")
    weightsPath := path.Join(currPath, "res10_300x300_ssd_iter_140000.caffemodel")
    //definedConfidence := 0.5
    //modelPath := "mask_detector.model"

    net := gocv.ReadNet(prototxtPath, weightsPath)

    //model, _ := tf.LoadSavedModel(modelPath, []string{"serve"}, nil)
    img := gocv.IMRead("/home/fz/Desktop/download.jpeg", gocv.IMReadUnchanged)

    //orig := img.CopyTo
    //imgSize := img.Size()

    img.ConvertTo(&img, gocv.MatTypeCV32F)
    imgBlob := gocv.BlobFromImage(img, 1.0, image.Point{X: 300, Y: 300}, gocv.NewScalar(104.0, 177.0, 123.0, 0), false, false)

    net.SetInput(imgBlob, "")
    detections := net.Forward("")

    //a, _ := detections.FromPtr(1,1,gocv.MatTypeCV32F,1,2)
    fmt.Printf("%v", detections.Total())

    //detectionsMat := detections.Reshape(1,1)
    //_, maxVal, _, maxLoc := gocv.MinMaxLoc(detectionsMat)
    //
    //fmt.Println(maxVal)
    //fmt.Println(maxLoc.X)

Answer 1

我使用下面这行代码成功地做到了这一点

detections := net.Forward("")

for i := 0; i < detections.Total(); i += 7 {
    confidence := detections.GetFloatAt(0, i+2)