We Keep Coding

Empty detections variable on darknet yolov4

I followed a video from Youtube about how to run Yolov4 using Darknet.
The thing is that when I run the program using the .exe file, it run perfectly with great FPS.
And now I have to change some part of the code to do the things I want, meaning that I have to run the code using its python code and not using the .exe file anymore.
But before any change, I tested to run it with no change but I only have the video running with LOW FPS and no bounding box shown up.
After some research I found that the "detections" variable is empty, and the problem comes on the "detect_image" function it self. The thing is that I can't figure out why it is empty even I know that there is something he should detect on it.
Here are the code from darknet_video.py
from ctypes import *
import random
import os
import cv2
import time
import darknet
import argparse
from threading import Thread, enumerate
from queue import Queue
def parser():
parser = argparse.ArgumentParser(description="YOLO Object Detection")
parser.add_argument("--input", type=str, default=0,
help="video source. If empty, uses webcam 0 stream")
parser.add_argument("--out_filename", type=str, default="",
help="inference video name. Not saved if empty")
parser.add_argument("--weights", default="yolov4.weights",
help="yolo weights path")
parser.add_argument("--dont_show", action='store_true',
help="windown inference display. For headless systems")
parser.add_argument("--ext_output", action='store_true',
help="display bbox coordinates of detected objects")
parser.add_argument("--config_file", default="./cfg/yolov4.cfg",
help="path to config file")
parser.add_argument("--data_file", default="./cfg/coco.data",
help="path to data file")
parser.add_argument("--thresh", type=float, default=.25,
help="remove detections with confidence below this value")
return parser.parse_args()
def str2int(video_path):
"""
argparse returns and string althout webcam uses int (0, 1 ...)
Cast to int if needed
"""
try:
return int(video_path)
except ValueError:
return video_path
def check_arguments_errors(args):
assert 0 < args.thresh < 1, "Threshold should be a float between zero and one (non-inclusive)"
if not os.path.exists(args.config_file):
raise(ValueError("Invalid config path {}".format(os.path.abspath(args.config_file))))
if not os.path.exists(args.weights):
raise(ValueError("Invalid weight path {}".format(os.path.abspath(args.weights))))
if not os.path.exists(args.data_file):
raise(ValueError("Invalid data file path {}".format(os.path.abspath(args.data_file))))
if str2int(args.input) == str and not os.path.exists(args.input):
raise(ValueError("Invalid video path {}".format(os.path.abspath(args.input))))
def set_saved_video(input_video, output_video, size):
fourcc = cv2.VideoWriter_fourcc(*"MJPG")
fps = int(input_video.get(cv2.CAP_PROP_FPS))
video = cv2.VideoWriter(output_video, fourcc, fps, size)
return video
def convert2relative(bbox):
"""
YOLO format use relative coordinates for annotation
"""
x, y, w, h = bbox
_height = darknet_height
_width = darknet_width
return x/_width, y/_height, w/_width, h/_height
def convert2original(image, bbox):
x, y, w, h = convert2relative(bbox)
image_h, image_w, __ = image.shape
orig_x = int(x * image_w)
orig_y = int(y * image_h)
orig_width = int(w * image_w)
orig_height = int(h * image_h)
bbox_converted = (orig_x, orig_y, orig_width, orig_height)
return bbox_converted
def convert4cropping(image, bbox):
x, y, w, h = convert2relative(bbox)
image_h, image_w, __ = image.shape
orig_left = int((x - w / 2.) * image_w)
orig_right = int((x + w / 2.) * image_w)
orig_top = int((y - h / 2.) * image_h)
orig_bottom = int((y + h / 2.) * image_h)
if (orig_left < 0): orig_left = 0
if (orig_right > image_w - 1): orig_right = image_w - 1
if (orig_top < 0): orig_top = 0
if (orig_bottom > image_h - 1): orig_bottom = image_h - 1
bbox_cropping = (orig_left, orig_top, orig_right, orig_bottom)
return bbox_cropping
def video_capture(frame_queue, darknet_image_queue):
while cap.isOpened():
ret, frame = cap.read()
if not ret:
break
frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
frame_resized = cv2.resize(frame_rgb, (darknet_width, darknet_height),
interpolation=cv2.INTER_LINEAR)
frame_queue.put(frame)
img_for_detect = darknet.make_image(darknet_width, darknet_height, 3)
darknet.copy_image_from_bytes(img_for_detect, frame_resized.tobytes())
darknet_image_queue.put(img_for_detect)
cap.release()
def inference(darknet_image_queue, detections_queue, fps_queue):
while cap.isOpened():
darknet_image = darknet_image_queue.get()
prev_time = time.time()
print("Network: ", network)
detections = darknet.detect_image(network, class_names, darknet_image, thresh=args.thresh)
detections_queue.put(detections)
fps = int(1/(time.time() - prev_time))
fps_queue.put(fps)
print("FPS: {}".format(fps))
darknet.print_detections(detections, args.ext_output)
darknet.free_image(darknet_image)
cap.release()
def drawing(frame_queue, detections_queue, fps_queue):
random.seed(3) # deterministic bbox colors
video = set_saved_video(cap, args.out_filename, (video_width, video_height))
while cap.isOpened():
frame = frame_queue.get()
detections = detections_queue.get()
fps = fps_queue.get()
detections_adjusted = []
if frame is not None:
for label, confidence, bbox in detections:
bbox_adjusted = convert2original(frame, bbox)
detections_adjusted.append((str(label), confidence, bbox_adjusted))
image = darknet.draw_boxes(detections_adjusted, frame, class_colors)
if len(detections):
name = []
count = []
for detect in detections:
name_tag = detect[0].decode()
if name_tag in name:
count[name.index(name_tag)] += 1
else:
name.append(name_tag)
count.append(1)
for index in len(name):
cv2.putText(image, name[index] + ": " + str(count[index]), (10, 30 + 30 * index), cv2.FONT_HERSHEY_SIMPLEX,0.5, (0, 0, 255), 2)
if not args.dont_show:
cv2.imshow('Inference', image)
if args.out_filename is not None:
video.write(image)
if cv2.waitKey(fps) == 27:
break
cap.release()
video.release()
cv2.destroyAllWindows()
if __name__ == '__main__':
frame_queue = Queue()
darknet_image_queue = Queue(maxsize=1)
detections_queue = Queue(maxsize=1)
fps_queue = Queue(maxsize=1)
args = parser()
check_arguments_errors(args)
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=1
)
darknet_width = darknet.network_width(network)
darknet_height = darknet.network_height(network)
input_path = str2int(args.input)
cap = cv2.VideoCapture(input_path)
video_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
video_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
Thread(target=video_capture, args=(frame_queue, darknet_image_queue)).start()
Thread(target=inference, args=(darknet_image_queue, detections_queue, fps_queue)).start()
Thread(target=drawing, args=(frame_queue, detections_queue, fps_queue)).start()
Thank you for helping

How to capture video by video from one rtsp Url using Opencv?

The server is sending video by video using the same RTSP URL(rtsp://192.168.0.2:8554/)
I can capture and display video using opencv.
import numpy as np
import cv2 as cv
os.environ["OPENCV_FFMPEG_CAPTURE_OPTIONS"] = "rtsp_transport;udp"
cap = cv.VideoCapture('rtsp://192.168.0.2:8554/')
while cap.isOpened():
ret, frame = cap.read()
# if frame is read correctly ret is True
if not ret:
print("Can't receive frame (stream end?). Exiting ...")
break
cv.imshow('frame', frame)
if cv.waitKey(1) == ord('q'):
break
cap.release()
cv.destroyAllWindows()
This program returns error when going on to the next video.
I tried this, but this didn't work.
import cv2 as cv
import os
import time
os.environ["OPENCV_FFMPEG_CAPTURE_OPTIONS"] = "rtsp_transport;udp"
cap = cv.VideoCapture('rtsp://192.168.0.26:8554/')
if not cap.isOpened():
print("Cannot open camera")
exit()
while True:
try:
time.sleep(2)
# Capture frame-by-frame
ret, frame = cap.read()
# if frame is read correctly ret is True
# Our operations on the frame come here
# Display the resulting frame
cv.imshow('frame',frame)
if cv.waitKey(1) == ord('q'):
break
except:
print("Exception!!")
# When everything done, release the capture
cap.release()
cv.destroyAllWindows()
Can I get some help?
Thanks in advance!

I solved this by using multi-threaded program.
Main file
from datasets import LoadStreams
import threading
import os
import logging
import cv2
import torch
import time
logger = logging.getLogger(__name__)
def select_device(device='', batch_size=None):
# device = 'cpu' or '0' or '0,1,2,3'
cpu_request = device.lower() == 'cpu'
if device and not cpu_request: # if device requested other than 'cpu'
os.environ['CUDA_VISIBLE_DEVICES'] = device # set environment variable
assert torch.cuda.is_available(), f'CUDA unavailable, invalid device {device} requested' # check availablity
cuda = False if cpu_request else torch.cuda.is_available()
if cuda:
c = 1024 ** 2 # bytes to MB
ng = torch.cuda.device_count()
if ng > 1 and batch_size: # check that batch_size is compatible with device_count
assert batch_size % ng == 0, f'batch-size {batch_size} not multiple of GPU count {ng}'
x = [torch.cuda.get_device_properties(i) for i in range(ng)]
s = f'Using torch {torch.__version__} '
for i, d in enumerate((device or '0').split(',')):
if i == 1:
s = ' ' * len(s)
logger.info(f"{s}CUDA:{d} ({x[i].name}, {x[i].total_memory / c}MB)")
else:
logger.info(f'Using torch {torch.__version__} CPU')
logger.info('') # skip a line
return torch.device('cuda:0' if cuda else 'cpu')
def detect(rtsp_url):
dataset = LoadStreams(rtsp_url)
device = select_device('')
count = 0
view_img = True
# img = torch.zeros((1, 3, imgsz, imgsz), device=device) # init img
try:
for frame_idx, (path, img, im0s, vid_cap) in enumerate(dataset): # for every frame
count += 1
im0 = im0s[0].copy()
if view_img:
cv2.imshow(str(path), im0)
# if cv2.waitKey(1) == ord('q'): # q to quit
# raise StopIteration
except:
print("finish execption")
dataset.stop()
return "good"
if __name__ == '__main__':
rtsp_url = "rtsp://192.168.0.26:8554/"
while True:
for thread in threading.enumerate():
print(thread.name)
print(detect(rtsp_url))
dataset class file
import glob
import logging
import math
import os
import random
import shutil
import time
import re
from itertools import repeat
from multiprocessing.pool import ThreadPool
from pathlib import Path
from threading import Thread
import cv2
import numpy as np
import torch
class LoadStreams: # multiple IP or RTSP cameras
def __init__(self, sources='streams.txt', img_size=640):
self.mode = 'stream'
self.img_size = img_size
self.capture = None
self.my_thread = None
self.stopFlag = False
if os.path.isfile(sources):
with open(sources, 'r') as f:
sources = [x.strip() for x in f.read().strip().splitlines() if len(x.strip())]
else:
sources = [sources]
n = len(sources)
self.imgs = [None] * n
self.sources = [clean_str(x) for x in sources] # clean source names for later
s = sources[0]
# for i, s in enumerate(sources):
# Start the thread to read frames from the video stream
# print('%g/%g: %s... ' % (i + 1, n, s), end='')
cap = cv2.VideoCapture(eval(s) if s.isnumeric() else s)
assert cap.isOpened(), 'Failed to open %s' % s
w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
fps = cap.get(cv2.CAP_PROP_FPS) % 100
self.ret, self.imgs[0] = cap.read() # guarantee first frame
thread = Thread(target=self.update, args=([0, cap]), daemon=True)
print(' success (%gx%g at %.2f FPS).' % (w, h, fps))
thread.start()
self.capture = cap
self.my_thread = thread
print('') # newline
# check for common shapes
s = np.stack([letterbox(x, new_shape=self.img_size)[0].shape for x in self.imgs], 0) # inference shapes
self.rect = np.unique(s, axis=0).shape[0] == 1 # rect inference if all shapes equal
if not self.rect:
print('WARNING: Different stream shapes detected. For optimal performance supply similarly-shaped streams.')
def update(self, index, cap):
# Read next stream frame in a daemon thread
n = 0
while cap.isOpened() and not self.stopFlag:
n += 1
# _, self.imgs[index] = cap.read()
cap.grab()
if n == 4: # read every 4th frame
_, self.imgs[index] = cap.retrieve()
n = 0
time.sleep(0.01) # wait time
def stop(self):
self.stopFlag = True
try:
# self.capture.release()
# self.my_thrsead.join()
print("stop thread!!")
except:
print("ERROR stopping thread!!")
def __iter__(self):
self.count = -1
return self
def __next__(self):
self.count += 1
img0 = self.imgs.copy()
if cv2.waitKey(1) == ord('q'): # q to quit
cv2.destroyAllWindows()
raise StopIteration
if not self.ret:
print("error!!!")
self.stop()
# Letterbox
img = [letterbox(x, new_shape=self.img_size, auto=self.rect)[0] for x in img0]
# Stack
img = np.stack(img, 0)
# Convert
img = img[:, :, :, ::-1].transpose(0, 3, 1, 2) # BGR to RGB, to bsx3x416x416
img = np.ascontiguousarray(img)
return self.sources, img, img0, None
def __len__(self):
return 0 # 1E12 frames = 32 streams at 30 FPS for 30 years
# def stop(self):
def clean_str(s):
# Cleans a string by replacing special characters with underscore _
return re.sub(pattern="[|##!¡·$€%&()=?¿^*;:,¨´><+]", repl="_", string=s)
def letterbox(img, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True):
# Resize image to a 32-pixel-multiple rectangle https://github.com/ultralytics/yolov3/issues/232
shape = img.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])
if not scaleup: # only scale down, do not scale up (for better test mAP)
r = min(r, 1.0)
# 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
if auto: # minimum rectangle
dw, dh = np.mod(dw, 32), np.mod(dh, 32) # wh padding
elif scaleFill: # stretch
dw, dh = 0.0, 0.0
new_unpad = (new_shape[1], new_shape[0])
ratio = new_shape[1] / shape[1], new_shape[0] / shape[0] # width, height ratios
dw /= 2 # divide padding into 2 sides
dh /= 2
if shape[::-1] != new_unpad: # resize
img = cv2.resize(img, 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))
img = cv2.copyMakeBorder(img, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color) # add border
return img, ratio, (dw, dh)
while cap.isOpened() and not self.stopFlag:
this line is especially important because
without this line the threads will be stacked and will have memory error
as the stack stacks up.

"No module named video" error in OpenCV-Python (Pycharm IDE)

My code:
import numpy as np
import cv2 as cv
# built-in modules
import sys
# local modules
import video
if __name__ == '__main__':
hsv_map = np.zeros((180, 256, 3), np.uint8)
h, s = np.indices(hsv_map.shape[:2])
hsv_map[:,:,0] = h
hsv_map[:,:,1] = s
hsv_map[:,:,2] = 255
hsv_map = cv.cvtColor(hsv_map, cv.COLOR_HSV2BGR)
cv.imshow('hsv_map', hsv_map)
cv.namedWindow('hist', 0)
hist_scale = 10
def set_scale(val):
global hist_scale
hist_scale = val
cv.createTrackbar('scale', 'hist', hist_scale, 32, set_scale)
try:
fn = sys.argv[1]
except:
fn = 0
cam = video.create_capture(fn,
fallback='synth:bg=../data/baboon.jpg:class=chess:noise=0.05')
while True:
flag, frame = cam.read()
cv.imshow('camera', frame)
small = cv.pyrDown(frame)
hsv = cv.cvtColor(small, cv.COLOR_BGR2HSV)
dark = hsv[...,2] < 32
hsv[dark] = 0
h = cv.calcHist([hsv], [0, 1], None, [180, 256], [0, 180, 0, 256])
h = np.clip(h*0.005*hist_scale, 0, 1)
vis = hsv_map*h[:,:,np.newaxis] / 255.0
cv.imshow('hist', vis)
ch = cv.waitKey(1)
if ch == 27:
break
cv.destroyAllWindows()
OpenCV - 3.4.1
This is the color histogram default code that comes with OpenCV-Python.
When I tried this code in Pycharm, it gives me "No module named video" message.
I tried some solutions from SOF and tried searching on the internet but I couldn't really solve or understand the problem properly.
Please help.
Thanks!

error occurred when using cv2.imshow()

The task is, i take pictures from the camera, and i process pictures using cnns, and then display the pic with the result. But there is an error when i use opencv python interface to show the image(the os is the mac os):
while(1):
test_data = []
ret, frame = cap.read()
frame = cv2.imread('hello.jpg')
h, w, _ = frame.shape
img = copy.deepcopy(frame)
img = pre_process(img)
test_data.append([img])
ret_res = _infer(inferer, test_data, threshold)
draw_result(frame, ret_res, h, w)
cv2.imshow("hellocapture", frame)
the following error occurred:
PC: # 0x0 (unknown)
*** SIGFPE (#0x7fffc5186d01) received by PID 36436 (TID 0x7fffe38773c0) stack
trace: ***
# 0x7fffdab7bb3a _sigtramp
# 0x7fffc5186d02 CFNumberCreate
# 0x7fffc6c027b5 -[NSPlaceholderNumber initWithDouble:]
# 0x7fffcae3b594 +[CALayer defaultValueForKey:]
# 0x7fffcaebd489 classDescription_locked()
# 0x7fffcaebc9fe classDescription_locked()
# 0x7fffcaebc9fe classDescription_locked()
# 0x7fffcaeb8292 classDescription()
# 0x7fffcaeb8495 CAObject_classInfo
# 0x7fffcae3bdf4 CA::Layer::class_state()
# 0x7fffcae3e081 -[CALayer init]
# 0x7fffc2d07854 -[NSView makeBackingLayer]
# 0x7fffc2d076db -[NSView(NSInternal) _createLayerAndInitialize]
# 0x7fffc2d06d4d -[NSView _doSetWantsLayerYES]
# 0x7fffc2d069da -[NSView setWantsLayer:]
# 0x7fffc2d142b6 __49-[NSThemeFrame _floatTitlebarAndToolbarFromInit:]_block_invoke
# 0x7fffc364b8b6 +[NSAnimationContext runAnimationGroup:]
# 0x7fffc2d13f23 -[NSThemeFrame _floatTitlebarAndToolbarFromInit:]
# 0x7fffc2d11a9c -[NSThemeFrame initWithFrame:styleMask:owner:]
# 0x7fffc2d10522 -[NSWindow _commonInitFrame:styleMask:backing:defer:]
# 0x7fffc2d0ec03 -[NSWindow _initContent:styleMask:backing:defer:contentView:]
# 0x7fffc2d0e65f -[NSWindow initWithContentRect:styleMask:backing:defer:]
# 0x1171ddebd -[QCocoaWindow initWithContentRect:styleMask:backing:defer:]
# 0x1171dddb1 -[NSWindow(QWidgetIntegration) qt_initWithQWidget:contentRect:styleMask:]
# 0x1171cee19 qt_mac_create_window()
# 0x1171ce119 QWidgetPrivate::createWindow_sys()
# 0x1171ce073 qt_mac_window_for()
# 0x1171d3b13 QWidgetPrivate::setModal_sys()
# 0x1172637c8 QWidget::create()
# 0x1175d5431 QToolBarPrivate::init()
# 0x1175d631c QToolBar::QToolBar()
# 0x111ca9c98 CvWindow::createToolBar()
Floating point exception: 8
and i've positioned that the cv2.imshow has the problem, but i don't know how?
if i just use the folloing code, it's ok
#!/usr/bin/env python
# coding=utf-8
import cv2
import numpy
import matplotlib.pyplot as plot
cap = cv2.VideoCapture(0)
while(1):
# get a frame
ret, frame = cap.read()
# show a frame
print frame.shape
cv2.imshow("capture", frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break

When you display an image in OpenCV you normally need to set a waitKey and release:
while(1):
test_data = []
ret, frame = cap.read()
frame = cv2.imread('hello.jpg')
h, w, _ = frame.shape
img = copy.deepcopy(frame)
img = pre_process(img)
test_data.append([img])
ret_res = _infer(inferer, test_data, threshold)
draw_result(frame, ret_res, h, w)
cv2.imshow("hellocapture", frame)
if cv2.waitKey(1) & 0xFF == ord('q')
break
cap.release()
cv2.destroyAllWindows()

Opencv python background substraction and motion tracking

I read about opencv on google and found the following sample code online to play with:
import cv2
def diffImg(t0, t1, t2):
d1 = cv2.absdiff(t2, t1)
d2 = cv2.absdiff(t1, t0)
return cv2.bitwise_and(d1, d2)
cam = cv2.VideoCapture('vid1.mp4')
winName = "Movement Indicator"
cv2.namedWindow(winName, cv2.CV_WINDOW_AUTOSIZE)
# Read three images first:
t_minus = cv2.cvtColor(cam.read()[1], cv2.COLOR_RGB2GRAY)
t = cv2.cvtColor(cam.read()[1], cv2.COLOR_RGB2GRAY)
t_plus = cv2.cvtColor(cam.read()[1], cv2.COLOR_RGB2GRAY)
while True:
cv2.imshow( winName, diffImg(t_minus, t, t_plus) )
# Read next image
t_minus = t
t = t_plus
t_plus = cv2.cvtColor(cam.read()[1], cv2.COLOR_RGB2GRAY)
key = cv2.waitKey(10)
if key == 27:
cv2.destroyWindow(winName)
break
print "Goodbye"
It produces the following kind of output for a sample video that I give to it:
So I am getting such results with this script. Now what I am trying to figure out is -
1) Get the bounding rect for the moving object in the video file
2) copy the contents of that bounding rect from original video frame to another and write the finished video to a file
Also, the screenshots on trying Kanishak Katahra's solution below I am getting the following output in the result window in the right side in the screenshot below -

Make the point of interest in the output you are getting as 1s. Make the rest as 0.
Convert it to 3D array of size same as original image.
Multiply original image with the output you are getting.
import cv2
import numpy as np
from numpy import newaxis
import time
def diffimg (a,b,c):
t0 = cv2.absdiff(a,b)
t1 = cv2.absdiff(b,c)
t3 = cv2.bitwise_and(t0,t1)
return t3
cap = cv2.VideoCapture(0)
t = cap.read() [1]
tp = cap.read() [1]
tpp = cap.read() [1]
t = cv2.cvtColor(t,cv2.COLOR_BGR2GRAY)
tp = cv2.cvtColor(tp,cv2.COLOR_BGR2GRAY)
tpp = cv2.cvtColor(tpp,cv2.COLOR_BGR2GRAY)
while True:
img = diffimg(t,tp,tpp)
cv2.imshow("motion detct",img)
key = cv2.waitKey(10)
res,img2 = cap.read()
#print img2.shape
t = tp
tp = tpp
tpp = cv2.cvtColor(img2,cv2.COLOR_BGR2GRAY)
cv2.imshow('image',img)
print img
img[img <= 10] = 0 #Try adjusting this value for better results.
img[img != 0] = 1
img = np.repeat(img[:, :, np.newaxis], 3, axis=2)
img3 = np.multiply( img,img2)
cv2.imshow('result',img3)
if key == 27:
cv2.destroyAllWindows()
break;
print "Goodbye User"