Yaxin
/
XZNSH-Code-AI
31
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update code

Browse Source
V0.1.0
王莹 2 years ago
parent 94bdfc7663
commit 0d19061d9a
2 changed files with 171 additions and 179 deletions
Show Stats Download Patch File Download Diff File

336
Bank_second_part/xbank_detect_process/main_process.py
Unescape Escape View File

@ -20,6 +20,8 @@ import json
def data_load(args):
# print('正在运行的进程',msg)
# print(args)
source = args[0]
model_ymal = args[1]
@ -36,217 +38,189 @@ def data_load(args):
if rtsp_source:
rtsp_detect_process(source=source, model_data=model_data,
model_inference=model_inference)
cap = cv2.VideoCapture(source)
if dir_source:
dir_source_process(source, model_inference, model_data)
if file_source:
# 视频流信息
fps = int(cap.get(cv2.CAP_PROP_FPS))
fps_num = fps*model_data['detect_time']
fps_num_small = fps*model_data['detect_time_small']
size = (int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)),
int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)))
try:
i = 0
j = 0
file_source_process(source, model_inference, model_data)
det_t_num = 0
nodet_t_num = 0
det_img = []
def rtsp_detect_process(source, model_data, model_inference):
video_name_time = 0
det_fps_time = []
cap = cv2.VideoCapture(source)
while True:
ret, frame = cap.read()
# 视频流信息
fps = int(cap.get(cv2.CAP_PROP_FPS))
fps_num = fps*model_data['detect_time']
fps_num_small = fps*model_data['detect_time_small']
size = (int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)),
int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)))
try:
i = 0
j = 0
if not ret:
continue # 如果未成功读取到视频帧，则继续读取下一帧
det_t_num = 0
nodet_t_num = 0
i = i + 1
j = j + 1
det_img = []
# 读取到当前视频帧时间
data_now = datetime.now()
get_time = str(data_now.strftime("%H")) + \
str(data_now.strftime("%M")) + str(data_now.strftime("%S")) + \
str(data_now.strftime("%f"))
video_name_time = 0
det_fps_time = []
# 视频保存
if video_name_time == 0:
while True:
ret, frame = cap.read()
video_name_time = get_time
savePath = os.path.join(model_data['save_videos'], (str(data_now.strftime(
"%Y")) + str(data_now.strftime("%m")) + str(data_now.strftime("%d"))))
t1 = time.time()
if not os.path.exists(savePath):
os.makedirs(savePath)
if not ret:
continue # 如果未成功读取到视频帧，则继续读取下一帧
video_path = os.path.join(
savePath, video_name_time + '.avi')
print('video_path:', video_path)
i = i + 1
j = j + 1
out_video = cv2.VideoWriter(
video_path, cv2.VideoWriter_fourcc(*'DIVX'), fps, size)
# 读取到当前视频帧时间
data_now = datetime.now()
get_time = str(data_now.strftime("%H")) + \
str(data_now.strftime("%M")) + str(data_now.strftime("%S")) + \
str(data_now.strftime("%f"))
print(source, data_now, i,j)
t1 = time.time()
# 视频保存
if video_name_time == 0:
imgframe_dict = {"path": source,
'frame': frame,
'get_fps': j}
video_name_time = get_time
savePath = os.path.join(model_data['save_videos'], (str(data_now.strftime(
"%Y")) + str(data_now.strftime("%m")) + str(data_now.strftime("%d"))))
images_det_result = img_process(imgframe_dict,
model_inference,
model_data)
if not os.path.exists(savePath):
os.makedirs(savePath)
images_update = save_process(imgframe_dict,
images_det_result,
model_data)
video_path = os.path.join(
savePath, video_name_time + '.avi')
print('video_path:', video_path)
print('images_det_result:',len(images_det_result))
out_video = cv2.VideoWriter(
video_path, cv2.VideoWriter_fourcc(*'DIVX'), fps, size)
if images_det_result:
print(source, data_now, i, j,video_path)
det_t_num = det_t_num + 1
# 推理部分
imgframe_dict = {"path": source, 'frame': frame, 'get_fps': j}
images_det_result = img_process(
imgframe_dict, model_inference, model_data)
# print(len(det_img))
if len(det_img) == 0:
img_dict = images_update.copy()
del img_dict['frame']
det_img.append(img_dict)
images_update = save_process(
imgframe_dict, images_det_result, model_data)
if not images_det_result and len(det_img) > 0:
# print('images_det_result:', len(images_det_result))
nodet_t_num = nodet_t_num + 1
# 结果判断，t
if images_det_result:
if (det_t_num + nodet_t_num) >= fps_num_small:
det_t_num = det_t_num + 1
para = determine_time(det_num=det_t_num,
nodet_num=nodet_t_num,
ratio_set=model_data['detect_ratio'])
if len(det_img) == 0:
img_dict = images_update.copy()
del img_dict['frame']
det_img.append(img_dict)
if para:
if not images_det_result and len(det_img) > 0:
nodet_t_num = nodet_t_num + 1
if (det_t_num + nodet_t_num) >= fps_num_small:
para = determine_time(
det_num=det_t_num, nodet_num=nodet_t_num, ratio_set=model_data['detect_ratio'])
first_fps_time = det_img[0]
print({"dert_fps": (j-int(first_fps_time['get_fps'])+1)})
first_fps_time.update({"dert_fps": (j-int(first_fps_time['get_fps'])+1)})
det_fps_time.append(first_fps_time)
if para:
det_img.clear()
det_t_num = 0
nodet_t_num = 0
first_fps_time = det_img[0]
first_fps_time.update(
{"dert_fps": (j-int(first_fps_time['get_fps'])+1)})
det_fps_time.append(first_fps_time)
# print('det_img:', len(det_img), det_t_num, nodet_t_num)
det_img.clear()
det_t_num = 0
nodet_t_num = 0
out_video.write(images_update['frame'])
# 视频保存
out_video.write(images_update['frame'])
if j >= fps_num:
# 结果判断 ，T
if j >= fps_num:
out_video.release()
if det_img:
out_video.release()
first_fps_time = det_img[0]
print({"dert_fps": (j-int(first_fps_time['get_fps'])+1)})
first_fps_time.update({"dert_fps": (j-int(first_fps_time['get_fps'])+1)})
det_fps_time.append(first_fps_time)
# T时间截至，判断t时间结果。
if det_img:
para = determine_time(
det_num=det_t_num, nodet_num=nodet_t_num, ratio_set=model_data['detect_ratio'])
print('det_fps_time:',det_fps_time)
first_fps_time = det_img[0]
if det_fps_time:
re_list = json_get(time_list=det_fps_time,video_path=video_path)
json_save(re_list)
# print(j-int(first_fps_time['get_fps'])+1)
# print(fps_num_small/2)
else:
print(video_path)
os.remove(video_path)
print('clear videos')
if (j-int(first_fps_time['get_fps'])+1) >= (fps_num_small/2):
first_fps_time.update(
{"dert_fps": (j-int(first_fps_time['get_fps'])+1)})
det_fps_time.append(first_fps_time)
det_fps_time.clear()
video_name_time = 0
j = 0
print('det_fps_time:', len(det_fps_time), i, j)
# break
t2 = time.time()
tx = t2 - t1
print('检测一张图片的时间为：', tx)
# print('det_fps_time:', det_fps_time)
except Exception as e:
# 处理异常或错误
print(str(e))
if det_fps_time:
re_list = json_get(
time_list=det_fps_time, video_path=video_path,fps=fps)
json_save(re_list)
cap.release()
else:
print(video_path)
os.remove(video_path)
print('----------------------------------------------clear videos-----------------------------------------------')
if dir_source:
# 重置
print('----------------------------------------------next-----------------------------------------------')
det_img.clear()
det_fps_time.clear()
det_t_num = 0
nodet_t_num = 0
video_name_time = 0
j = 0
img_ext = [".jpg", ".JPG", ".bmp"]
video_ext = [".mp4", ".avi", ".MP4"]
# print('det_fps_time:', det_fps_time,'det_img:',det_img)
img_list = get_dir_file(source, img_ext)
video_list = get_dir_file(source, video_ext)
t2 = time.time()
tx = t2 - t1
print('检测一张图片的时间为：', tx)
if img_list:
except Exception as e:
# 处理异常或错误
print(str(e))
for img in img_list:
cap.release()
t1 = time.time()
images = cv2.imread(img)
imgframe_dict = {"path": img, 'frame': images}
def dir_source_process(source, model_inference, model_data):
images_update = img_process(
imgframe_dict, model_inference, model_data)
img_ext = [".jpg", ".JPG", ".bmp"]
video_ext = [".mp4", ".avi", ".MP4"]
t2 = time.time()
tx = t2 - t1
print('检测一张图片的时间为：', tx)
img_list = get_dir_file(source, img_ext)
video_list = get_dir_file(source, video_ext)
if video_list:
if img_list:
pass
for img in img_list:
if file_source:
t1 = time.time()
images = cv2.imread(img)
img_para = True
imgframe_dict = {"path": img, 'frame': images}
if img_para:
images = cv2.imread(source)
imgframe_dict = {"path": source, 'frame': images}
images_update = img_process(
imgframe_dict, model_inference, model_data)
t2 = time.time()
tx = t2 - t1
print('检测一张图片的时间为：', tx)
if video_list:
pass
def file_source_process(source, model_inference, model_data):
img_para = True
if img_para:
images = cv2.imread(source)
imgframe_dict = {"path": source, 'frame': images}
images_update = img_process(
imgframe_dict, model_inference, model_data)
def img_process(images, model_inference, model_data):
@ -273,15 +247,17 @@ def img_process(images, model_inference, model_data):
else:
determine_bbox = select_labels_list
# print(determine_bbox)
if model_data['model_parameter']['object_num_min']:
if len(determine_bbox) >= model_data["model_parameter"]['object_num_min']:
print(determine_bbox)
if model_data['model_parameter']['object_num_min'] :
if len(determine_bbox) <= model_data["model_parameter"]['object_num_min']:
# print(len(determine_bbox))
print(len(determine_bbox))
determine_bbox.clear()
# 返回检测后结果
return determine_bbox
@ -347,6 +323,9 @@ def images_save(images, save_path):
# data_now.year) + '/' + str(data_now.month) + '_' + str(data_now.day) + '/'
img_save_path = os.path.join(save_path, str(images['path'].split('.')[-1]))
images_name = images['get_time'] + '.jpg'
# img_save_path = save_path + '/' + str(images['path'].split('.')[-1]) + '/'
# print(img_save_path)
if not os.path.exists(img_save_path):
os.makedirs(img_save_path)
@ -381,7 +360,7 @@ def determine_time(det_num, nodet_num, ratio_set):
ratio = det_num / (det_num + nodet_num)
# print(det_num, nodet_num, ratio)
print(det_num, nodet_num, ratio)
if ratio >= ratio_set:
@ -409,14 +388,13 @@ def video_synthesis(imglist, savePath, size, fps, videoname):
out.release()
def json_get(time_list, video_path,fps):
def json_get(time_list,video_path):
result_dict ={'info': {'video_path': video_path,'fps':fps}}
re_dict = {}
for i, det_dict in enumerate(time_list):
result_dict = {'video_path':video_path}
for i,det_dict in enumerate(time_list):
list_hands = ["Keypad", "hands", "keyboard", "mouse", "phone"]
list_sleep = ["person", "sleep"]
list_hands = ["Keypad","hands","keyboard", "mouse","phone"]
list_sleep = ["person","sleep"]
if list(det_dict['results'][0].keys())[0] in list_hands:
@ -426,22 +404,36 @@ def json_get(time_list, video_path,fps):
result_lables = "sleep"
fps_dict = {'time': det_dict['get_fps'],
'duration': det_dict['dert_fps'], 'result': result_lables}
re_dict.update({('id_' + str(i)): fps_dict})
fps_dict = {'time': det_dict['get_fps'],'duration':det_dict['dert_fps'],'result':result_lables}
result_dict.update({('id_'+ str(i)):fps_dict})
result_dict.update({'result':re_dict})
return result_dict
# def json_analysis(re_list):
# update_list = []
# copy_list = [x for x in re_list not in update_list]
# for i in range(len(copy_list)-1):
# j = i + 1
# re_i = int(re_list[i]['fps'])
# re_i_add = int(re_list[i]['dert_fps'])
# re_j = int(re_list[j]['fps'])
# if re_i + re_i_add == re_j:
# update_list.append(re_i,re_j)
# print()
def json_save(result_dict):
json_path = result_dict['info`']['video_path'].split('.')[0] + '.json'
result = json.dumps(result_dict)
json_path = result_dict['video_path'].split('.')[0] + '.json'
del result_dict['video_path']
result = json.dumps(result_dict)
f = open(json_path, 'w')
f = open(json_path,'w')
f.write(result + '\n')
f.close

14
Bank_second_part/xbank_detect_process/model_load/model_load.py
Unescape Escape View File

@ -9,13 +9,13 @@ def build_option(device, backend, cache_file):
"""
option = fd.RuntimeOption()
option.use_cpu()
option.trt_option.serialize_file = cache_file
if device.lower() == "gpu":
option.use_gpu(0)
if backend.lower() == "trt":
assert device.lower(
) == "gpu", "TensorRT backend require inference on device GPU."
option.use_trt_backend()
# option.trt_option.serialize_file = cache_file
# if device.lower() == "gpu":
# option.use_gpu(0)
# if backend.lower() == "trt":
# assert device.lower(
# ) == "gpu", "TensorRT backend require inference on device GPU."
# option.use_trt_backend()
return option

Write Preview
Loading…
Cancel
Save