目录
板卡信息简介
表 板卡信息
序号 | 项目 | 备注 |
操作系统 | 银河麒麟 V10 | 内核为linux 5.10.160 |
芯片 | ARM64 | RK3588 |
接口 | u口、网口 | |
电压 | 9~24V | |
librknnrt | 1.5.2 |
摄像头信息简介
海康摄像头(200万像素,4mm定焦镜头,具有补光能力):
图 相机尺寸
表 技术参数
-
数据接入方法
安装ros系统,下载rocon_devices-develop
网址: GitHub - robotics-in-concert/rocon_devices: A collection of device drivers involved in rocon
具体步骤:
mkdir -p ~/catkin_ws/src
cd ~/catkin_ws/src
catkin_init_workspace
在src/rocon_devices-develop/rocon_rtsp_camera_relay/launch/rtsp_camera_relay.launch中,修改 第二行 default="rtsp://用户名(例:amdin):密码(例:A12345)@ 摄像头IP与端口号(例:192.168.254.12:554)/Streaming/Channels/101"
将rocon_devices-develop添加到src文件夹下:
输入catkin_make
启动命令 roslaunch rocon_rtsp_camera_relay rtsp_camera_relay.launch
输出topic名称:/rtsp_camera_relay/image
-
部分识别代码(python)
#!/usr/bin/env python3
import sys
sys.path.append('/opt/ros/kinetic/lib/python2.7/dist-packages')
sys.path.append('/usr/lib/python2.7/dist-packages')
import rospy
import time
from sensor_msgs.msg import Image
from cv_bridge import CvBridge,CvBridgeError
import numpy as np
import sys
python3 = True if sys.hexversion > 0x03000000 else False
if python3:
# remove ros python2 library
ros_cv_path = '/opt/ros/kinetic/lib/python2.7/dist-packages'
if ros_cv_path in sys.path:
sys.path.remove(ros_cv_path)
rospy.init_node('listener', anonymous=True)
import cv2
import os
import time
import copy
from pathlib import Path
import platform
from rknnlite.api import RKNNLite
#----------------------------------------------------
# decice tree for rk356x/rk3588
DEVICE_COMPATIBLE_NODE = '/proc/device-tree/compatible'
img = []
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'
else:
host = 'RK356x'
except IOError:
print('Read device node {} failed.'.format(DEVICE_COMPATIBLE_NODE))
exit(-1)
else:
host = os_machine
return host
RK356X_RKNN_MODEL = 'resnet18_for_rk356x.rknn'
RK3588_RKNN_MODEL = '/home/kylin/桌面/newfolder/Yolov5_rknnlite2-main/weights/yolov5s-640-640.rknn'
OBJ_THRESH = 0.25
NMS_THRESH = 0.45
IMG_SIZE = 640
CLASSES = ("person",)
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):
b1, c1, s1 = process(input, mask, anchors)
b1, c1, s1 = filter_boxes(b1, c1, s1)
boxes.append(b1)
classes.append(c1)
scores.append(s1)
boxes = np.concatenate(boxes)
boxes = xywh2xyxy(boxes)
classes = np.concatenate(classes)
scores = np.concatenate(scores)
nboxes, nclasses, nscores = [], [], []
for c in set(classes):
if c != 0:
continue
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)
#---------------------------------------------------
cv_type = dict()
cv_type['mono8'] = (np.uint8 , 1)
cv_type['8UC2'] = (np.uint8 , 2)
cv_type['bgr8'] = (np.uint8 , 3)
cv_type['8UC4'] = (np.uint8 , 4)
cv_type['8SC1'] = (np.int8 , 1)
cv_type['8SC2'] = (np.int8 , 2)
cv_type['8SC3'] = (np.int8 , 3)
cv_type['8SC4'] = (np.int8 , 4)
cv_type['16UC1'] = (np.uint16 , 1)
cv_type['16UC2'] = (np.uint16 , 2)
cv_type['16UC3'] = (np.uint16 , 3)
cv_type['16UC4'] = (np.uint16 , 4)
cv_type['16SC1'] = (np.int16 , 1)
cv_type['16SC2'] = (np.int16 , 2)
cv_type['16SC3'] = (np.int16 , 3)
cv_type['16SC4'] = (np.int16 , 4)
cv_type['32SC1'] = (np.uint32 , 1)
cv_type['32SC2'] = (np.uint32 , 2)
cv_type['32SC3'] = (np.uint32 , 3)
cv_type['32SC4'] = (np.uint32 , 4)
cv_type['32FC1'] = (np.int32 , 1)
cv_type['32FC2'] = (np.int32 , 2)
cv_type['32FC3'] = (np.int32 , 3)
cv_type['32FC4'] = (np.int32 , 4)
cv_type['64FC1'] = (np.float64, 1)
cv_type['64FC2'] = (np.float64, 2)
cv_type['64FC3'] = (np.float64, 3)
cv_type['64FC4'] = (np.float64, 4)
ros_type = dict()
ros_type[('uint8' , 1)] = ('mono8', 1)
ros_type[('uint8' , 2)] = ('8UC2' , 2)
ros_type[('uint8' , 3)] = ('bgr8' , 3)
ros_type[('uint8' , 4)] = ('8UC4' , 4)
ros_type[('int8' , 1)] = ('8SC1' , 1)
ros_type[('int8' , 2)] = ('8SC2' , 2)
ros_type[('int8' , 3)] = ('8SC3' , 3)
ros_type[('int8' , 4)] = ('8SC4' , 4)
ros_type[('uint16' , 1)] = ('16UC1', 1)
ros_type[('uint16' , 2)] = ('16UC2', 2)
ros_type[('uint16' , 3)] = ('16UC3', 3)
ros_type[('uint16' , 4)] = ('16UC4', 4)
ros_type[('int16' , 1)] = ('16SC1', 1)
ros_type[('int16' , 2)] = ('16SC2', 2)
ros_type[('int16' , 3)] = ('16SC3', 3)
ros_type[('int16' , 4)] = ('16SC4', 4)
ros_type[('uint32' , 1)] = ('32SC1', 1)
ros_type[('uint32' , 2)] = ('32SC2', 2)
ros_type[('uint32' , 3)] = ('32SC3', 3)
ros_type[('uint32' , 4)] = ('32SC4', 4)
ros_type[('int32' , 1)] = ('32FC1', 1)
ros_type[('int32' , 2)] = ('32FC2', 2)
ros_type[('int32' , 3)] = ('32FC3', 3)
ros_type[('int32' , 4)] = ('32FC4', 4)
ros_type[('float64', 1)] = ('64FC1', 1)
ros_type[('float64', 2)] = ('64FC2', 2)
ros_type[('float64', 3)] = ('64FC3', 3)
ros_type[('float64', 4)] = ('64FC4', 4)
def ToRosMsg(img):
msg = Image()
msg.header.stamp = rospy.Time.now()
msg.width = img.shape[1]
msg.height = img.shape[0]
channel = 1
if len(img.shape) is 3:
channel = img.shape[2]
msg.encoding = ros_type[(str(img.dtype), channel)][0]
msg.is_bigendian = 0
msg.step = msg.width * channel
msg.data = img.tobytes()
return msg
def ToCvImg(msg):
img = np.fromstring(
msg.data, dtype=cv_type[msg.encoding][0]).reshape(
msg.height, msg.width, cv_type[msg.encoding][1])
return img
def callback(imgmsg):
img = ToCvImg(imgmsg)
img = cv2.resize(img,(640,416),interpolation=cv2.INTER_NEAREST)
host_name = get_host()
if host_name == 'RK356x':
rknn_model = RK356X_RKNN_MODEL
elif 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':
ret = rknn_lite.init_runtime(core_mask=RKNNLite.NPU_CORE_0)
else:
ret = rknn_lite.init_runtime()
if ret != 0:
print('Init runtime environment failed')
exit(ret)
print('done')
img, ratio, (dw, dh) = letterbox(img, new_shape=(IMG_SIZE, IMG_SIZE))
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
print('--> Running model')
outputs = rknn_lite.inference(inputs=[img])
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)
if boxes is not None:
draw(img_1, boxes, scores, classes)
print('-----------')
cv2.imshow("camera", img_1)
cv2.waitKey(1)
#cv2.imshow("listener", img)
else:
cv2.imshow("camera", img_1)
cv2.waitKey(1)
rknn_lite.release()
def listener():
rospy.Subscriber("/rtsp_camera_relay/image", Image, callback)
rospy.spin()
if __name__ == '__main__':
listener()
识别结果:
优点,方法识别延时较小,长时间运行较少出现卡顿,死机等情况。