import os
import urllib
import traceback
import time
import sys
import numpy as np
import cv2
# from rknnpool import rknnPoolExecutor
import platform
from rknnlite.api import RKNNLite
from rknnpool import rknnPoolExecutor
from func import myFunc
# decice tree for RK356x/RK3588
DEVICE_COMPATIBLE_NODE = '/proc/device-tree/compatible'
RK3588_RKNN_MODEL = '/home/tyzc/RK_NPU_SDK_1.5.0/rknn-toolkit2/examples/onnx/yolov5/yolov5s_640_640_cut.rknn'
IMG_PATH = '/home/tyzc/RK_NPU_SDK_1.5.0/rknn-toolkit2/examples/onnx/yolov5/bus.jpg'
DATASET = '/home/tyzc/RK_NPU_SDK_1.5.0/rknn-toolkit2/examples/onnx/yolov5/dataset.txt'
OBJ_THRESH = 0.25
NMS_THRESH = 0.45
IMG_SIZE_HEIGHT = 640
IMG_SIZE_WEIGHT = 640
IMG_SIZE = 640
CLASSES = ("person", "cartoon")
video_path = "/home/tyzc/0416_syl_chaofen/rknn-multi-threaded-Super-Resolution-syl/cartorn/1怀旧动漫社聪明的一休_360_640.mp4"
cap = cv2.VideoCapture(video_path)
modelPath = "/home/tyzc/0416_syl_chaofen/rknn-multi-threaded-Super-Resolution-syl/rknnModel/32_1_16_4channel_123conv_1rdb1_net_oneresize_no_convhr_360_320_20240407_36.0778.rknn"
modelPath_1 = "/home/tyzc/0416_syl_chaofen/rknn-multi-threaded-Super-Resolution-syl/rknnModel/mix_0925/mix_32_1_16_4channel_123conv_1rdb1_net_oneresize_no_convhr_360_320_31.15..rknn"
# modelPath_1 = "/home/tyzc/0416_syl_chaofen/rknn-multi-threaded-Super-Resolution-syl/rknnModel/32_1_16_4channel_123conv_1rdb1_net_oneresize_no_convhr_360_320_20240407_36.0778.rknn"
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
print("视频分辨率:{} x {}".format(height, width))
# modelPath = "/home/tyzc/Downloads/rknn-toolkit2-master/rknn-multi-threaded-nosigmoid/rknnModel/8_1_4/8_1_4_net_g_160000_Y_360_640_doqu.rknn"
# 线程数, 增大可提高帧率
TPEs = 6
# 初始化rknn池
pool = rknnPoolExecutor(
rknnModel=modelPath,
TPEs=TPEs,
func=myFunc)
pool_1 = rknnPoolExecutor(
rknnModel=modelPath,
TPEs=TPEs,
func=myFunc)
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])
box_confidence = sigmoid(input[..., 4])
box_confidence = np.expand_dims(box_confidence, axis=-1)
box_class_probs = sigmoid(input[..., 5:])
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)
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)
grid = np.concatenate((col, row), axis=-1)
box_xy += grid
box_xy *= int(IMG_SIZE/grid_h)
box_wh = pow(sigmoid(input[..., 2:4])*2, 2)
box_wh = box_wh * anchors
box = np.concatenate((box_xy, box_wh), axis=-1)
return box, box_confidence, box_class_probs
def filter_boxes(boxes, box_confidences, box_class_probs):
"""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.
"""
boxes = boxes.reshape(-1, 4)
box_confidences = box_confidences.reshape(-1)
box_class_probs = box_class_probs.reshape(-1, box_class_probs.shape[-1])
_box_pos = np.where(box_confidences >= OBJ_THRESH)
boxes = boxes[_box_pos]
box_confidences = box_confidences[_box_pos]
box_class_probs = box_class_probs[_box_pos]
class_max_score = np.max(box_class_probs, axis=-1)
classes = np.argmax(box_class_probs, axis=-1)
_class_pos = np.where(class_max_score >= OBJ_THRESH)
boxes = boxes[_class_pos]
classes = classes[_class_pos]
scores = (class_max_score* box_confidences)[_class_pos]
return boxes, classes, scores
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 = [], [], []
for input, mask in zip(input_data, masks):
b, c, s = process(input, mask, anchors)
b, c, s = filter_boxes(b, c, s)
boxes.append(b)
classes.append(c)
scores.append(s)
boxes = np.concatenate(boxes)
boxes = xywh2xyxy(boxes)
classes = np.concatenate(classes)
scores = np.concatenate(scores)
nboxes, nclasses, nscores = [], [], []
for c in set(classes):
inds = np.where(classes == c)
b = boxes[inds]
c = classes[inds]
s = scores[inds]
keep = nms_boxes(b, s)
nboxes.append(b[keep])
nclasses.append(c[keep])
nscores.append(s[keep])
if not nclasses and not nscores:
return None, None, None
boxes = np.concatenate(nboxes)
classes = np.concatenate(nclasses)
scores = np.concatenate(nscores)
return boxes, classes, scores
def draw(image, boxes, scores, classes):
"""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 in zip(boxes, scores, classes):
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)
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)
def get_host():
# get platform and device type
system = platform.system()
machine = platform.machine()
os_machine = system + '-' + machine
if os_machine == 'Linux-aarch64':
try:
with open(DEVICE_COMPATIBLE_NODE) as f:
device_compatible_str = f.read()
if 'rk3588' in device_compatible_str:
host = 'RK3588'
elif 'rk3562' in device_compatible_str:
host = 'RK3562'
else:
host = 'RK3566_RK3568'
except IOError:
print('Read device node {} failed.'.format(DEVICE_COMPATIBLE_NODE))
exit(-1)
else:
host = os_machine
return host
if __name__ == '__main__':
# Get device information
host_name = get_host()
if host_name == 'RK3588':
rknn_model = RK3588_RKNN_MODEL
else:
print("This demo cannot run on the current platform: {}".format(host_name))
exit(-1)
rknn_lite = RKNNLite()
# Load RKNN model
print('--> Load RKNN model')
ret = rknn_lite.load_rknn(rknn_model)
if ret != 0:
print('Load RKNN model failed')
exit(ret)
print('done')
# Init runtime environment
print('--> Init runtime environment')
# run on RK356x/RK3588 with Debian OS, do not need specify target.
if host_name == 'RK3588':
# For RK3588, specify which NPU core the model runs on through the core_mask parameter.
ret = rknn_lite.init_runtime(core_mask=RKNNLite.NPU_CORE_0_1_2)
# ret = rknn_lite.init_runtime()
else:
ret = rknn_lite.init_runtime()
if ret != 0:
print('Init runtime environment failed')
exit(ret)
print('done')
frames, loopTime, initTime = 0, time.time(), time.time()
all_read_time = 0
all_show_time = 0
person_num = 0
cartoon_num = 0
while (cap.isOpened()):
if person_num + cartoon_num <= 10:
real_time_start = time.time()
# all_read_time = all_read_time + (read_time_2 - read_time_1)
frames += 1
ret, frame = cap.read()
frame_h, frame_w, c = frame.shape
if not ret:
break
# Set inputs
img = frame
# 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_lite.inference(inputs=[img], data_format=['nchw'])
inference_time_start = time.time()
outputs = rknn_lite.inference(inputs=[img])
inference_time_end = time.time()
print("-" * 60)
print(img.shape)
print(inference_time_end - inference_time_start)
# post process
input0_data = outputs[0]
input1_data = outputs[1]
input2_data = outputs[2]
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 = yolov5_post_process(input_data)
img_1 = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
# 统计0和1的数目
if boxes is not None:
count_0 = np.count_nonzero(classes == 0)
count_1 = np.count_nonzero(classes == 1)
person_num = person_num + count_0
cartoon_num = cartoon_num + count_1
draw(img_1, boxes, scores, classes)
# cv2.imwrite('result.jpg', img_1)
# 释放资源
show_time_1 = time.time()
# 设置文字、字体、大小、颜色等
text = "person_num: " + str(person_num) + " " + 'cartoon_num: ' + str(cartoon_num)
org = (50, 100) # 文字放置的坐标 (x, y)
font = cv2.FONT_HERSHEY_SIMPLEX # 字体类型
font_scale = 1.0 # 字体大小
color = (0, 0, 255) # 文字颜色,白色 (B, G, R)
thickness = 2 # 文字线条粗细
frame_resize_show = cv2.resize(frame, (frame.shape[1] * 2, frame.shape[0] * 2), 2, 2)
# 在图片上写文字
cv2.putText(frame_resize_show, text, org, font, font_scale, color, thickness)
cv2.imshow('test', frame_resize_show)
show_time_2 = time.time()
if cv2.waitKey(1) & 0xFF == ord('q'):
# out.release()
break
print("person_num: " + str(person_num) + " " + 'cartoon_num: ' + str(cartoon_num))
real_time_end = time.time()
print("-"*24)
print(real_time_end - real_time_start)
else:
rknn_lite.release()
if cartoon_num > person_num:
# 初始化异步所需要的帧
pool_frames = []
if (cap.isOpened()):
for i in range(TPEs + 1):
ret, frame = cap.read()
frame_cf = frame[:,:320,]
frame_resize = frame[:,320:,]
if not ret:
cap.release()
del pool
exit(-1)
pool_frames.append(frame_resize)
pool.put(frame_cf)
while True:
frames += 1
ret, frame = cap.read()
if not ret:
break
frame_resize = frame[:,:320,]
frame_cf = frame[:,320:,]
read_time_2 = time.time()
if frames % 100000000 == 0:
# 调整图像大小
frame = cv2.resize(frame, (frame.shape[1] * 2, frame.shape[0] * 2), 2, 2)
else:
pool.put(frame_cf)
pool_frames.append(frame_resize)
frame_cf, flag = pool.get()
frame_resize = cv2.resize(pool_frames.pop(0), (frame.shape[1], frame.shape[0] * 2), 2, 2)
# 创建一个空白帧,大小与输入视频相同
result_frame = np.zeros((height*2, width*2, 3), dtype=np.uint8)
# 将左半部分放置在左边
result_frame[:, :width] = frame_resize
# 将右半部分放置在右边
result_frame[:, width:] = frame_cf
result_frame[:, width:width+1, :] = (255, 255, 255)
frame = result_frame
if flag == False:
break
show_time_1 = time.time()
# 设置文字、字体、大小、颜色等
text = "is_cartoon"
org = (50, 100) # 文字放置的坐标 (x, y)
font = cv2.FONT_HERSHEY_SIMPLEX # 字体类型
font_scale = 1.0 # 字体大小
color = (0, 0, 255) # 文字颜色,白色 (B, G, R)
thickness = 2 # 文字线条粗细
# 在图片上写文字
cv2.putText(frame, text, org, font, font_scale, color, thickness)
cv2.imshow('test', frame)
show_time_2 = time.time()
if cv2.waitKey(1) & 0xFF == ord('q'):
# out.release()
break
if frames % 30 == 0:
print("30帧平均帧率:\t", 30 / (time.time() - loopTime - all_read_time - all_show_time), "帧")
loopTime = time.time()
all_read_time = 0
all_show_time = 0
elif person_num > cartoon_num:
# 初始化异步所需要的帧
pool_frames = []
if (cap.isOpened()):
for i in range(TPEs + 1):
ret, frame = cap.read()
frame_cf = frame[:,:320,]
frame_resize = frame[:,320:,]
if not ret:
cap.release()
del pool
exit(-1)
pool_frames.append(frame_resize)
pool_1.put(frame_cf)
while True:
frames += 1
ret, frame = cap.read()
if not ret:
break
frame_resize = frame[:,:320,]
frame_cf = frame[:,320:,]
read_time_2 = time.time()
if frames % 100000000 == 0:
# 调整图像大小
frame = cv2.resize(frame, (frame.shape[1] * 2, frame.shape[0] * 2), 2, 2)
else:
pool_1.put(frame_cf)
pool_frames.append(frame_resize)
frame_cf, flag = pool_1.get()
frame_resize = cv2.resize(pool_frames.pop(0), (frame.shape[1], frame.shape[0] * 2), 2, 2)
# 创建一个空白帧,大小与输入视频相同
result_frame = np.zeros((height*2, width*2, 3), dtype=np.uint8)
# 将左半部分放置在左边
result_frame[:, :width] = frame_resize
# 将右半部分放置在右边
result_frame[:, width:] = frame_cf
result_frame[:, width:width+1, :] = (255, 255, 255)
frame = result_frame
if flag == False:
break
show_time_1 = time.time()
# 设置文字、字体、大小、颜色等
text = "is_person"
org = (50, 100) # 文字放置的坐标 (x, y)
font = cv2.FONT_HERSHEY_SIMPLEX # 字体类型
font_scale = 1.0 # 字体大小
color = (0, 0, 255) # 文字颜色,白色 (B, G, R)
thickness = 2 # 文字线条粗细
# 在图片上写文字
cv2.putText(frame, text, org, font, font_scale, color, thickness)
cv2.imshow('test', frame)
show_time_2 = time.time()
if cv2.waitKey(1) & 0xFF == ord('q'):
# out.release()
break
if frames % 30 == 0:
print("30帧平均帧率:\t", 30 / (time.time() - loopTime - all_read_time - all_show_time), "帧")
loopTime = time.time()
all_read_time = 0
all_show_time = 0
print("总平均帧率\t", frames / (time.time() - initTime))
# 释放cap和rknn线程池
cap.release()
# out.release()
cv2.destroyAllWindows()
pool_1.release()
pool.release()
03-04
2443
08-10
3251