如何在android中向tflite模型传递图像

Question

我已经将一个Yolo模型转换为.tflite，以便在Android中使用。在python中就是这样使用的--

net = cv2.dnn.readNet("yolov2.weights", "yolov2.cfg")
classes = []
with open("yolov3.txt", "r") as f:
    classes = [line.strip() for line in f.readlines()]
layer_names = net.getLayerNames()
output_layers = [layer_names[i[0] - 1] for i in net.getUnconnectedOutLayers()]
colors = np.random.uniform(0, 255, size=(len(classes), 3))

cap= cv2.VideoCapture(0)


while True:
    _,frame= cap.read()
    height,width,channel= frame.shape
    blob = cv2.dnn.blobFromImage(frame, 0.00392, (320, 320), (0, 0, 0), True, crop=False)
    net.setInput(blob)
    outs = net.forward(output_layers)
    for out in outs:
        for detection in out:
            scores = detection[5:]
            class_id = np.argmax(scores)
            confidence = scores[class_id]
            if confidence > 0.2:
            # Object detected
                center_x = int(detection[0] * width)
                center_y = int(detection[1] * height)
                w = int(detection[2] * width)
                h = int(detection[3] * height)
                # Rectangle coordinates
                x = int(center_x - w / 2)
                y = int(center_y - h / 2)

我使用netron https://github.com/lutzroeder/netron可视化模型。输入描述为name:inputs，类型:float32[1,416,416,3]，quantization:0≤q≤255，位置:399，输出描述为name:output_boxs，类型:float32[1,106 47,8]，位置：400。

Answer 1

您需要调整输入图像的大小，使其与TensorFlow-Lite模型的输入大小相匹配，然后将其转换为RGB格式以馈送到模型。

通过使用TensorFlow-Lite支持库中的ImageProcessor，您可以轻松地进行图像大小调整和转换。

ImageProcessor imageProcessor =
        new ImageProcessor.Builder()
            .add(new ResizeWithCropOrPadOp(cropSize, cropSize))
            .add(new ResizeOp(imageSizeX, imageSizeY, ResizeMethod.NEAREST_NEIGHBOR))
            .add(new Rot90Op(numRoration))
            .add(getPreprocessNormalizeOp())
            .build();
return imageProcessor.process(inputImageBuffer);

接下来使用解释器运行推断，您将预处理后的图像提供给TensorFlow-Lite解释器：

tflite.run(inputImageBuffer.getBuffer(), outputProbabilityBuffer.getBuffer().rewind());

请参考这个官方示例了解更多详细信息，此外，您也可以参考这个示例。