Body detection added

nsfw_video_file.py

@@ -0,0 +1,96 @@
+import cv2
+from keras.models import load_model
+import numpy as np
+from collections import deque
+import warnings
+warnings.filterwarnings("ignore")
+
+model = load_model("/content/drive/My Drive/nsfw/Final_weights.h5")
+
+labels = {0 : "Neutral", 1 : "Porn", 2 : "Sexy"}
+
+size = 128
+# input_vid = "2.mp4"
+output_vid = "1.mp4"
+
+Q = deque(maxlen=size)
+
+
+
+
+
+
+
+# vs = cv2.VideoCapture(0)
+vs = cv2.VideoCapture('/content/drive/My Drive/nsfw/3.mp4')
+writer = None
+(W, H) = (None, None)
+
+# loop over frames from the video file stream
+while True:
+    # read the next frame from the file
+    (grabbed, frame) = vs.read()
+
+    # if the frame was not grabbed, then we have reached the end
+    # of the stream
+    if not grabbed:
+        break
+
+    # if the frame dimensions are empty, grab them
+    if W is None or H is None:
+        (H, W) = frame.shape[:2]
+    fc = cv2.CascadeClassifier("drive/My Drive/nsfw/haarcascade_upperbody.xml")
+    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)    
+    upper=fc.detectMultiScale(gray,1.1,1)
+
+    for (a,b,c,d) in upper:
+      cv2.rectangle(frame,(a,b),(a+c,b+d),(0,0,255),2)
+
+
+
+    output = frame.copy()
+    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+    frame = frame/255.0
+    frame = cv2.resize(frame, (224, 224)).astype("float32")
+
+#     frame -= mean
+
+    # make predictions on the frame and then update the predictions
+    # queue
+    preds = model.predict(np.expand_dims(frame, axis=0))[0]
+    print(preds)
+    Q.append(preds)
+
+    # perform prediction averaging over the current history of
+    # previous predictions
+
+    results = np.array(Q).mean(axis=0)
+    i = np.argmax(preds)
+    label = labels[i]
+    # draw the activity on the output frame
+    text = "activity: {}:".format(label)
+    cv2.putText(output, text, (35, 50), cv2.FONT_HERSHEY_SIMPLEX, 1.25, (0, 255, 0), 5)
+
+    # check if the video writer is None
+    if writer is None:
+        # initialize our video writer
+        fourcc = cv2.VideoWriter_fourcc(*"MJPG")
+        writer = cv2.VideoWriter(output_vid, fourcc, 30, (W, H), True)
+
+    # write the output frame to disk
+    writer.write(output)
+
+    # show the output image
+    # cv2_imshow(output)
+    key = cv2.waitKey(1) & 0xFF
+
+    # if the `q` key was pressed, break from the loop
+    if key == ord("q"):
+        break
+
+# release the file pointers
+print("[INFO] cleaning up...")
+# writer.release()
+vs.release()
+
+

nsfw_video_realtime.py

@@ -0,0 +1,92 @@
+import cv2
+from keras.models import load_model
+import numpy as np
+from collections import deque
+import warnings
+warnings.filterwarnings("ignore")
+
+model = load_model("/content/drive/My Drive/nsfw/Final_weights.h5")
+
+labels = {0 : "Neutral", 1 : "Porn", 2 : "Sexy"}
+
+
+size = 128
+# input_vid = "2.mp4"
+output_vid = "1.mp4"
+
+Q = deque(maxlen=size)
+
+
+from google.colab.patches import cv2_imshow
+import imutils
+
+# vs = cv2.VideoCapture(0)
+vs = cv2.VideoCapture(take_photo())
+writer = None
+(W, H) = (None, None)
+
+# loop over frames from the video file stream
+while True:
+    # read the next frame from the file
+    (grabbed, frame) = vs.read()
+
+    # if the frame was not grabbed, then we have reached the end
+    # of the stream
+    if not grabbed:
+        break
+
+    # if the frame dimensions are empty, grab them
+    if W is None or H is None:
+        (H, W) = frame.shape[:2]
+
+    # frame=imutils.resize(frame, width=min(100, frame.shape[1])) 
+    fc = cv2.CascadeClassifier("drive/My Drive/nsfw/haarcascade_upperbody.xml")
+    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)    
+    upper=fc.detectMultiScale(gray,1.1,1)
+
+    for (a,b,c,d) in upper:
+      cv2.rectangle(frame,(a,b),(a+c,b+d),(0,0,255),2)
+    output = frame.copy()
+    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+    frame = frame/255.0
+    frame = cv2.resize(frame, (224, 224)).astype("float32")
+
+#     frame -= mean
+
+    # make predictions on the frame and then update the predictions
+    # queue
+    preds = model.predict(np.expand_dims(frame, axis=0))[0]
+    print(preds)
+    Q.append(preds)
+
+    # perform prediction averaging over the current history of
+    # previous predictions
+
+    results = np.array(Q).mean(axis=0)
+    i = np.argmax(preds)
+    label = labels[i]
+    # draw the activity on the output frame
+    text = "activity: {}:".format(label)
+    cv2.putText(output, text, (35, 50), cv2.FONT_HERSHEY_SIMPLEX, 1.25, (0, 255, 0), 5)
+
+    # check if the video writer is None
+    if writer is None:
+        # initialize our video writer
+        fourcc = cv2.VideoWriter_fourcc(*"MJPG")
+        writer = cv2.VideoWriter(output_vid, fourcc, 30, (W, H), True)
+
+    # write the output frame to disk
+    writer.write(output)
+
+    # show the output image
+    cv2_imshow(output)
+    key = cv2.waitKey(1) & 0xFF
+
+    # if the `q` key was pressed, break from the loop
+    if key == ord("q"):
+        break
+
+# release the file pointers
+print("[INFO] cleaning up...")
+# writer.release()
+vs.release()