rknn-toolkit推理yolov-face.rknn模型

import os
import urllib
import traceback
import time
import sys
import numpy as np
import cv2
from rknn.api import RKNN


ONNX_MODEL = 'yolov5s.onnx'
RKNN_MODEL = 'yolov5s.rknn'
IMG_PATH = './bus.jpg'
DATASET = './dataset.txt'

QUANTIZE_ON = True

BOX_THRESH = 0.5
NMS_THRESH = 0.4
IMG_SIZE = 640

#CLASSES = ("face")


CLASSES = ("person", "bicycle", "car","motorbike ","aeroplane ","bus ","train","truck ","boat","traffic light",
           "fire hydrant","stop sign ","parking meter","bench","bird","cat","dog ","horse ","sheep","cow","elephant",
           "bear","zebra ","giraffe","backpack","umbrella","handbag","tie","suitcase","frisbee","skis","snowboard","sports ball","kite",
           "baseball bat","baseball glove","skateboard","surfboard","tennis racket","bottle","wine glass","cup","fork","knife ",
           "spoon","bowl","banana","apple","sandwich","orange","broccoli","carrot","hot dog","pizza ","donut","cake","chair","sofa",
           "pottedplant","bed","diningtable","toilet ","tvmonitor","laptop	","mouse	","remote ","keyboard ","cell phone","microwave ",
           "oven ","toaster","sink","refrigerator ","book","clock","vase","scissors ","teddy bear ","hair drier", "toothbrush ")


def sigmoid(x):
    return 1 / (1 + np.exp(-x))

def xywh2xyxy(x):
    # Convert [x, y, w, h] to [x1, y1, x2, y2]
    y = np.copy(x)
    y[:, 0] = x[:, 0] - x[:, 2] / 2  # top left x
    y[:, 1] = x[:, 1] - x[:, 3] / 2  # top left y
    y[:, 2] = x[:, 0] + x[:, 2] / 2  # bottom right x
    y[:, 3] = x[:, 1] + x[:, 3] / 2  # bottom right y
    return y

def process(input, mask, anchors):

    anchors = [anchors[i] for i in mask]
    grid_h, grid_w = map(int, input.shape[0:2])
    print("grind_h::", grid_h)
    print("grind_w::", grid_w)#这里就是20*20, 40*40, 80*80;

    box_confidence = sigmoid(input[..., 4])
    box_confidence = np.expand_dims(box_confidence, axis=-1)

    #box_class_probs = sigmoid(input[..., 5:])
    box_class_probs = sigmoid(input[..., 15:])

    box_xy = sigmoid(input[..., :2])*2 - 0.5

    col = np.tile(np.arange(0, grid_w), grid_w).reshape(-1, grid_w)
    row = np.tile(np.arange(0, grid_h).reshape(-1, 1), grid_h)
    print("col after tile:", col)
    print("row after tile:", row)#这里的col row是0 1 2 3 4 5....20；
    col = col.reshape(grid_h, grid_w, 1, 1).repeat(3, axis=-2)
    row = row.reshape(grid_h, grid_w, 1, 1).repeat(3, axis=-2)
    print("col after reshape:", col)
    print("row after reshape:", row)
    grid = np.concatenate((col, row), axis=-1)#[col, row]
    print("grid after concatenate::", grid)
    box_xy += grid #[x+col, y+row]
    box_xy *= int(IMG_SIZE/grid_h) #相当于C++里面的那个*stride。

    box_wh = pow(sigmoid(input[..., 2:4])*2, 2)#pow表示x的y次方。
    print("box_wh::", box_wh)
    box_wh = box_wh * anchors
    print("box_xy:::", box_xy)

    box = np.concatenate((box_xy, box_wh), axis=-1)
    print("input[..., 2:4]::", input[..., 5:15])
    print("anchors:::",anchors)

    lanmark1_xy = (input[..., 5:7]*anchors    +  grid)*(int(IMG_SIZE/grid_h))
    lanmark2_xy = (input[..., 7:9]*anchors    +  grid)*(int(IMG_SIZE/grid_h))
    lanmark3_xy = (input[..., 9:11]*anchors   +  grid)*(int(IMG_SIZE/grid_h))
    lanmark4_xy = (input[..., 11:13]*anchors  +  grid)*(int(IMG_SIZE/grid_h))
    lanmark5_xy = (input[..., 13:15]*anchors  +  grid)*(int(IMG_SIZE/grid_h))
    
    print("lanmark1_xy:", lanmark1_xy)
    print("lanmark2_xy:", lanmark2_xy)
    print("lanmark3_xy:", lanmark3_xy)
    print("lanmark4_xy:", lanmark4_xy)
    print("lanmark5_xy:", lanmark5_xy)
    
    lanmarks = np.concatenate((lanmark1_xy, lanmark2_xy, lanmark3_xy, lanmark4_xy, lanmark5_xy), axis=-1)

    return box, box_confidence, box_class_probs, lanmarks

def filter_boxes(boxes, box_confidences, box_class_probs, landmarks):
    """Filter boxes with box threshold. It's a bit different with origin yolov5 post process!

    # Arguments
        boxes: ndarray, boxes of objects.
        box_confidences: ndarray, confidences of objects.
        box_class_probs: ndarray, class_probs of objects.

    # Returns
        boxes: ndarray, filtered boxes.
        classes: ndarray, classes for boxes.
        scores: ndarray, scores for boxes.
    """
    box_classes = np.argmax(box_class_probs, axis=-1)
    box_class_scores = np.max(box_class_probs, axis=-1)
    pos = np.where(box_confidences[...,0] >= BOX_THRESH)


    boxes = boxes[pos]
    classes = box_classes[pos]
    scores = box_class_scores[pos]
    marks = landmarks[pos]

    return boxes, classes, scores, marks

def nms_boxes(boxes, scores):
    """Suppress non-maximal boxes.

    # Arguments
        boxes: ndarray, boxes of objects.
        scores: ndarray, scores of objects.

    # Returns
        keep: ndarray, index of effective boxes.
    """
    x = boxes[:, 0]
    y = boxes[:, 1]
    w = boxes[:, 2] - boxes[:, 0]
    h = boxes[:, 3] - boxes[:, 1]

    areas = w * h
    order = scores.argsort()[::-1]

    keep = []
    while order.size > 0:
        i = order[0]
        keep.append(i)

        xx1 = np.maximum(x[i], x[order[1:]])
        yy1 = np.maximum(y[i], y[order[1:]])
        xx2 = np.minimum(x[i] + w[i], x[order[1:]] + w[order[1:]])
        yy2 = np.minimum(y[i] + h[i], y[order[1:]] + h[order[1:]])

        w1 = np.maximum(0.0, xx2 - xx1 + 0.00001)
        h1 = np.maximum(0.0, yy2 - yy1 + 0.00001)
        inter = w1 * h1

        ovr = inter / (areas[i] + areas[order[1:]] - inter)
        inds = np.where(ovr <= NMS_THRESH)[0]
        order = order[inds + 1]
    keep = np.array(keep)
    return keep


def yolov5_post_process(input_data):
    masks = [[0, 1, 2], [3, 4, 5], [6, 7, 8]]
    anchors = [[10, 13], [16, 30], [33, 23], [30, 61], [62, 45],
              [59, 119], [116, 90], [156, 198], [373, 326]]

    boxes, classes, scores, landmarks = [], [], [], []
    for input,mask in zip(input_data, masks):
        b, c, s, l = process(input, mask, anchors)
        b, c, s, l = filter_boxes(b, c, s, l)
        boxes.append(b)
        classes.append(c)
        scores.append(s)
        landmarks.append(l)


    boxes = np.concatenate(boxes)
    boxes = xywh2xyxy(boxes)
    classes = np.concatenate(classes)
    scores = np.concatenate(scores)
    landmarks = np.concatenate(landmarks)


    nboxes, nclasses, nscores, nlandmarks = [], [], [], []
    for c in set(classes):
        inds = np.where(classes == c)
        b = boxes[inds]
        c = classes[inds]
        s = scores[inds]
        l = landmarks[inds]

        keep = nms_boxes(b, s)

        nboxes.append(b[keep])
        nclasses.append(c[keep])
        nscores.append(s[keep])
        nlandmarks.append(l[keep])

    if not nclasses and not nscores:
        return None, None, None

    boxes = np.concatenate(nboxes)
    classes = np.concatenate(nclasses)
    scores = np.concatenate(nscores)
    landmarks = np.concatenate(nlandmarks)

    return boxes, classes, scores, landmarks

def draw(image, boxes, scores, classes, landmarks):
    """Draw the boxes on the image.

    # Argument:
        image: original image.
        boxes: ndarray, boxes of objects.
        classes: ndarray, classes of objects.
        scores: ndarray, scores of objects.
        all_classes: all classes name.
    """
    for box, score, cl, landmark in zip(boxes, scores, classes, landmarks):
        top, left, right, bottom = box
        print('class: {}, score: {}'.format(CLASSES[cl], score))
        print('box coordinate left,top,right,down: [{}, {}, {}, {}]'.format(top, left, right, bottom))
        top = int(top)
        left = int(left)
        right = int(right)
        bottom = int(bottom)
        print("(landmark[0], landmark[1])...:", landmark[0], landmark[1])
        cv2.circle(image, (int(landmark[0]), int(landmark[1])), 3, (255,0,0),0)
        cv2.circle(image, (int(landmark[2]), int(landmark[3])), 3, (255,0,0),0)
        cv2.circle(image, (int(landmark[4]), int(landmark[5])), 3, (255,0,0),0)
        cv2.circle(image, (int(landmark[6]), int(landmark[7])), 3, (255,0,0),0)
        cv2.circle(image, (int(landmark[8]), int(landmark[9])), 3, (255,0,0),0)

    

        cv2.rectangle(image, (top, left), (right, bottom), (255, 0, 0), 2)
        cv2.putText(image, '{0} {1:.2f}'.format(CLASSES[cl], score),
                    (top, left - 6),
                    cv2.FONT_HERSHEY_SIMPLEX,
                    0.6, (0, 0, 255), 2)


def letterbox(im, new_shape=(640, 640), color=(0, 0, 0)):
    # Resize and pad image while meeting stride-multiple constraints
    shape = im.shape[:2]  # current shape [height, width]
    if isinstance(new_shape, int):
        new_shape = (new_shape, new_shape)

    # Scale ratio (new / old)
    r = min(new_shape[0] / shape[0], new_shape[1] / shape[1])

    # Compute padding
    ratio = r, r  # width, height ratios
    new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r))
    dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1]  # wh padding

    dw /= 2  # divide padding into 2 sides
    dh /= 2

    if shape[::-1] != new_unpad:  # resize
        im = cv2.resize(im, new_unpad, interpolation=cv2.INTER_LINEAR)
    top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
    left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
    im = cv2.copyMakeBorder(im, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color)  # add border
    return im, ratio, (dw, dh)


if __name__ == '__main__':

    # Create RKNN object
    rknn = RKNN()
    

    # Load rknn model 
    ret = rknn.load_rknn("./yolov5s-face.rknn") 
    if ret != 0: 
           print('Load RKNN model failed.') 
           exit(ret)


    # init runtime environment
    print('--> Init runtime environment')
    ret = rknn.init_runtime()
    # ret = rknn.init_runtime('rk1808', device_id='1808')
    if ret != 0:
        print('Init runtime environment failed')
        exit(ret)
    print('done')

    # Set inputs
    img = cv2.imread("ldh.jpeg")
    # img, ratio, (dw, dh) = letterbox(img, new_shape=(IMG_SIZE, IMG_SIZE))
    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    img = cv2.resize(img,(IMG_SIZE, IMG_SIZE))

    # Inference
    print('--> Running model')
    outputs = rknn.inference(inputs=[img])

    # post process
    input0_data = outputs[0]
    input1_data = outputs[1]
    input2_data = outputs[2]
    
    print("input0_data::", input0_data.shape)

    input0_data = input0_data.reshape([3,-1]+list(input0_data.shape[-2:]))
    input1_data = input1_data.reshape([3,-1]+list(input1_data.shape[-2:]))
    input2_data = input2_data.reshape([3,-1]+list(input2_data.shape[-2:]))

    input_data = list()
    input_data.append(np.transpose(input0_data, (2, 3, 0, 1)))
    input_data.append(np.transpose(input1_data, (2, 3, 0, 1)))
    input_data.append(np.transpose(input2_data, (2, 3, 0, 1)))

    boxes, classes, scores, landmarks = yolov5_post_process(input_data)
    print("classes:", classes)
    print("landmarks::::", landmarks)
    img_1 = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
    if boxes is not None:
        draw(img_1, boxes, scores, classes, landmarks)
    #cv2.imshow("post process result", img_1)
    cv2.imwrite("./result.jpg", img_1)
    #cv2.waitKeyEx(0)

    rknn.release()