Develop Reference

How to record audio while OpenCv has enabled camera

I want to record the audio and store the same while analyzing the emotion using Opencv. But unfortunately, the audio which is being recorded is of minimal length and the voice is also not clear. I want to analyze the voice emotion too for which I need to audio. Could somebody help me in resolving the same?
def main():
cap = cv2.VideoCapture(0)
cap.set(cv2.CAP_PROP_FRAME_WIDTH, 1280)
cap.set(cv2.CAP_PROP_FRAME_HEIGHT, 720)
p = pyaudio.PyAudio()
stream = p.open(format=audio_format, channels=channels,rate=16000, input=True,frames_per_buffer=1024)
start_time = time.time()
aud = True
while aud:
ret, frame = cap.read()
data = stream.read(chunk, exception_on_overflow = False)
frames.append(data)
if not ret:
break
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors= 5, flags=cv2.CASCADE_SCALE_IMAGE)
result = DeepFace.analyze(img_path = frame , actions = ["emotion"], enforce_detection=False,detector_backend='ssd')
for (x,y,w,h) in faces:
if w > 130: #trick: ignore small faces
draw_border(frame, (x, y), (x + w, y + h), (255, 0, 105),4, 15, 10) ## draw rectangle around face.
detected_face = frame[int(y):int(y+h), int(x):int(x+w)] #crop detected face
detected_face = cv2.cvtColor(detected_face, cv2.COLOR_BGR2GRAY) #transform to gray scale
detected_face = cv2.resize(detected_face, (48, 48)) #resize to 48x48
img_pixels = img_to_array(detected_face)
img_pixels = np.expand_dims(img_pixels, axis = 0)
img_pixels /= 255 #pixels are in scale of [0, 255]. normalize all pixels in scale of [0, 1]
emotion = result["dominant_emotion"]
txt = str(emotion)
cv2.putText(frame,txt,(50,50),cv2.FONT_HERSHEY_SIMPLEX,1,(0,0,255),3)
cv2.imshow(file, frame)
if cv2.waitKey(1) & 0xFF == ord("q"):
aud=False
break
stream.stop_stream()
stream.close()
p.terminate()
wf = wave.open('/Users/xyz/Documents/Audio/wv.wav', 'wb')
wf.setnchannels(1)
wf.setsampwidth(p.get_sample_size(audio_format))
wf.setframerate(sample_rate)
wf.writeframes(b''.join(frames))
wf.close()
cap.release()
cv2.destroyAllWindows()

Pose Correction Dataset

i wanted to work on pose correction system for my college final year project and needed a my own datasets. but i am not able to get the pose coordinates from the video into CSV file no matter what i try. I tried but there is no code or reference which can help me in it.
So can anyone please tell how should i do it.
this is my code
import cv2
mp_drawing = mp.solutions.drawing_utils
mp_pose = mp.solutions.pose
cap = cv2.VideoCapture('C:/Users/lenovo/Downloads/pexels-julia-larson-6455077.mp4')
count=0
## Setup mediapipe instance
with mp_pose.Pose(min_detection_confidence=0.5, min_tracking_confidence=0.5) as pose:
while cap.isOpened():
ret, frame = cap.read()
# Recolor image to RGB
image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
image.flags.writeable = False
# Make detection
results = pose.process(image)
# Recolor back to BGR
image.flags.writeable = True
image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
# Extract landmarks
try:
landmarks = results.pose_landmarks.landmark
LANDMARKS = landmarks[mp_pose.PoseLandmark.value]
df = pd.DataFrame(LANDMARKS,columns)
df.to_csv (r'C:/Users/lenovo/Downloads/dataset4.csv', index = False, header=True)
except:
pass
# Render detections
mp_drawing.draw_landmarks(image, results.pose_landmarks, mp_pose.POSE_CONNECTIONS,
mp_drawing.DrawingSpec(color=(245,117,66), thickness=2, circle_radius=2),
mp_drawing.DrawingSpec(color=(245,66,230), thickness=2, circle_radius=2)
)
cv2.imshow('Mediapipe Feed', image)
if cv2.waitKey(10) & 0xFF == ord('q'):
break
cap.release()
cv2.destroyAllWindows()

Try moving the lines saving the data to pandas DataFrame outside the while loop.
Create an empty list: landmarks_list
For each frame append landmarks to list.
Run the followning lines after the while has finished running over the video frames
df = pd.DataFrame(landmarks_list,columns)
df.to_csv (r'C:/Users/lenovo/Downloads/dataset4.csv', index = False, header=True)

Tensorflow Object detection API: Print detected class as output

I am using the TF Object detection API to detect images, it is working fine and given an image it will draw the bounding box with a label and confidence score. My question is how to print the detected class (as a string) i.e not just on the image but as an output to the terminal too.
This is the code of detection in real time.
cap = cv2.VideoCapture(0)
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
while True:
ret, frame = cap.read()
image_np = np.array(frame)
input_tensor = tf.convert_to_tensor(np.expand_dims(image_np, 0),
dtype=tf.float32)
detections = detect_fn(input_tensor)
num_detections = int(detections.pop('num_detections'))
detections = {key: value[0, :num_detections].numpy()
for key, value in detections.items()}
detections['num_detections'] = num_detections
# detection_classes should be ints.
detections['detection_classes'] = detections['detection_classes'].astype(np.int64)
label_id_offset = 1
image_np_with_detections = image_np.copy()
viz_utils.visualize_boxes_and_labels_on_image_array(
image_np_with_detections,
detections['detection_boxes'],
detections['detection_classes']+label_id_offset,
detections['detection_scores'],
category_index,
use_normalized_coordinates=True,
max_boxes_to_draw=3,
min_score_thresh=.5,
agnostic_mode=False)
cv2.imshow('object detection', cv2.resize(image_np_with_detections, (800, 600)))
if cv2.waitKey(1) & 0xFF == ord('q'):
cap.release()
break

Classes will be encrypted in category_index variable.Use the below code snippet to get the detected class.
print ([category_index.get(value) for index,value in enumerate(classes[0]) if scores[0,index] > 0.5])

Code:
my_classes = detections['detection_classes'][0].numpy() + label_id_offset
my_scores = detections['detection_scores'][0].numpy()
min_score = 0.5
print([category_index[value]['name']
for index,value in enumerate(my_classes)
if my_scores[index] > min_score
])
Sample output:
['person', 'cell phone', 'remote']

How to draw rectangle in opencv around fire detected object from keras model

This code is for detecting fire using tensorflow keras model and I do not know how to draw rectangle around the fire detected object:
model = tf.keras.models.load_model('my_model\\InceptionV3.h5')
video = cv2.VideoCapture(0)
while True:
_, frame = video.read()
im = Image.fromarray(frame, 'RGB')
im = im.resize((224, 224))
img_array = image.img_to_array(im)
img_array = np.expand_dims(img_array, axis=0) / 255
probabilities = model.predict(img_array)[0]
prediction = np.argmax(probabilities)
# if prediction is 0, which means there is fire in the frame.
if prediction == 0:
img = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
print(probabilities[prediction])
cv2.imshow("Capturing", frame)
cv2.imshow("Capturing", img)
if cv2.waitKey(1) & 0xff == ord('q'):
break
video.release()
cv2.destroyAllWindows()

Convert Numpy Array to Video(mp4)

I have a video(mp4) that I read into a Numpy Array and edit. Now I want to convert the array back to a video
I already tried to use VideoWriter but the created mp4 was always corrupt
frame_array = []
with detection_graph.as_default():
with tf.Session(graph=detection_graph) as sess:
while cap.isOpened():
ret, image_np = cap.read()
if ret == True:
image_np_expanded = np.expand_dims(image_np, axis=0)
image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
scores = detection_graph.get_tensor_by_name('detection_scores:0')
classes = detection_graph.get_tensor_by_name('detection_classes:0')
num_detections = detection_graph.get_tensor_by_name('num_detections:0')
(boxes, scores, classes, num_detections) = sess.run(
[boxes, scores, classes, num_detections],
feed_dict={image_tensor: image_np_expanded})
vis_util.visualize_boxes_and_labels_on_image_array(
image_np,
np.squeeze(boxes),
np.squeeze(classes).astype(np.int32),
np.squeeze(scores),
category_index,
use_normalized_coordinates=True,
line_thickness=8)
i=i+1
frame_array.append(image_np)
cv2.imwrite('pictures/'+ str(i) + '.jpg',image_np)
else:
break
print("Everything detected")
out = cv2.VideoWriter('video.mp4',cv2.VideoWriter_fourcc(*'DIVX'), 30, (640, 480))
for i in frame_array:
out.write(i)
out.release()
The created mp4 was always corrupt and couldnt be opened. Every saved Picture with cv2.imwrite is okay and viewable