def _draw_faces(self, image, faces):
"""Take an Image message and a Faces message, draw a boc around the
faces in the image and publish the result.
Record *image*'s sequence number in _last_published_seq.
"""
# Convert image into RGB image
rgb_im = image_to_array(image)
# Use OpenCV to draw boxes in imahe
for face in faces.faces:
roi = face.roi
# NB: cv2.rectangle will modify the image it works on.
cv2.rectangle(rgb_im,
(roi.x_offset, roi.y_offset),
(roi.x_offset + roi.width, roi.y_offset + roi.height),
(0, 255, 0), # green
2, # thickness
)
# Publish image
self._image_pub.publish(array_to_image(rgb_im))
self._last_published_seq = image.header.seq
def _draw_faces(self, image, faces):
"""Take an Image message and a Faces message, draw a boc around the
faces in the image and publish the result.
Record *image*'s sequence number in _last_published_seq.
"""
# Convert image into RGB image
rgb_im = image_to_array(image)
# Use OpenCV to draw boxes in imahe
for face in faces.faces:
roi = face.roi
# NB: cv2.rectangle will modify the image it works on.
cv2.rectangle(
rgb_im,
(roi.x_offset, roi.y_offset),
(roi.x_offset + roi.width, roi.y_offset + roi.height),
(0, 255, 0), # green
2, # thickness
)
# Publish image
self._image_pub.publish(array_to_image(rgb_im))
self._last_published_seq = image.header.seq
def _detect_faces_callback(self, event):
"""Called periodically to detect faces in an image and publish the
result.
"""
# Get the latest image to arrive
image = self._latest_image
# Don't do anything if there is no latest image
if image is None:
return
# "Claim" this image by clearing _latest_image
self._latest_image = None
# Create the face message we will eventually publish
faces_msg = Faces()
# Copy the header from the input image so we know when and what these
# face detection results relate to.
faces_msg.header = image.header
# Parse image message into numpy array
image_array = image_to_array(image)
rospy.logdebug('Successfully parsed new image with shape: %s',
image_array.shape)
# Convert image to grayscale
gray_im = cv2.cvtColor(image_array, cv2.COLOR_RGB2GRAY)
# Detect faces
faces = self._detect_faces(gray_im)
# Create a face bounding box for each face
for face_idx, face_bbox in enumerate(faces):
rospy.logdebug('Face #%s at %s', face_idx + 1, face_bbox)
# Create face message
face_msg = Face()
face_msg.header = faces_msg.header
# Initialise roi
face_msg.roi.x_offset = face_bbox[0]
face_msg.roi.y_offset = face_bbox[1]
face_msg.roi.width = face_bbox[2]
face_msg.roi.height = face_bbox[3]
face_msg.roi.do_rectify = False
# Append to list of faces
faces_msg.faces.append(face_msg)
# Publish faces messages
self._faces_pub.publish(faces_msg)
def _detect_faces_callback(self, event):
"""Called periodically to detect faces in an image and publish the
result.
"""
# Get the latest image to arrive
image = self._latest_image
# Don't do anything if there is no latest image
if image is None:
return
# "Claim" this image by clearing _latest_image
self._latest_image = None
# Create the face message we will eventually publish
faces_msg = Faces()
# Copy the header from the input image so we know when and what these
# face detection results relate to.
faces_msg.header = image.header
# Parse image message into numpy array
image_array = image_to_array(image)
rospy.logdebug('Successfully parsed new image with shape: %s',
image_array.shape)
# Convert image to grayscale
gray_im = cv2.cvtColor(image_array, cv2.COLOR_RGB2GRAY)
# Detect faces
faces = self._detect_faces(gray_im)
# Create a face bounding box for each face
for face_idx, face_bbox in enumerate(faces):
rospy.logdebug('Face #%s at %s', face_idx+1, face_bbox)
# Create face message
face_msg = Face()
face_msg.header = faces_msg.header
# Initialise roi
face_msg.roi.x_offset = face_bbox[0]
face_msg.roi.y_offset = face_bbox[1]
face_msg.roi.width = face_bbox[2]
face_msg.roi.height = face_bbox[3]
face_msg.roi.do_rectify = False
# Append to list of faces
faces_msg.faces.append(face_msg)
# Publish faces messages
self._faces_pub.publish(faces_msg)