Skip to content

Instantly share code, notes, and snippets.

@OlafenwaMoses
Created September 29, 2021 09:59
Show Gist options
  • Save OlafenwaMoses/4a47d274a6c58de4965be3ff962ca106 to your computer and use it in GitHub Desktop.
Save OlafenwaMoses/4a47d274a6c58de4965be3ff962ca106 to your computer and use it in GitHub Desktop.
Sample code to detect objects in IP Camera with DeepStack
import cv2
# Import DeepStack's Python SDK
from deepstack_sdk import ServerConfig, Detection
# Function to draw detections and object names on camera frames
def draw_detections(img, detections):
for detection in detections:
output_font_scale = 0.8e-3 * img.shape[0]
label = detection.label
img = cv2.rectangle(
img,
(detection.x_min, detection.y_min),
(detection.x_max, detection.y_max),
(0,146,224),
2
)
img = cv2.putText(
img=img,
text=label + " ( " + str(100*detection.confidence)+"% )",
org=(detection.x_min-10, detection.y_min-10),
fontFace=cv2.FONT_HERSHEY_SIMPLEX,
fontScale=output_font_scale,
color=(0,146,224),
thickness=2
)
return img
if __name__=="__main__":
# Initiate Connection to DeepStack
config = ServerConfig("http://localhost:80")
detection = Detection(config)
# Initiate Connection to iPhone IP Camera
# Make sure you CHANGE THE IP to the one displayed on your iPhone
capture = cv2.VideoCapture('http://admin:password@192.168.1.230:8081/video')
while(True):
# Capture the video frame
# by frame
ret, frame = capture.read()
if ret:
# Detect the Frame with DeepStack using the Python SDK
detections = detection.detectObject(frame,output=None)
# Draw the detections on the frame
frame = draw_detections(frame, detections)
#Display the frame and the detections
cv2.imshow('frame', frame)
# the 'q' button is set as the
# quitting button you may use any
# desired button of your choice
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# After the loop release the cap detectionect
capture.release()
# Destroy all the windows
cv2.destroyAllWindows()
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment