Add hand tracking python src

pysrc/hand.py

+import cv2
+import mediapipe as mp
+# shared memory
+import mmap
+import struct
+
+# 랜드마크를 그리기 위한 변수
+mp_drawing = mp.solutions.drawing_utils
+# 손을 처리하기 위한 변수
+mp_drawing_styles = mp.solutions.drawing_styles
+# 랜드마크 표시 스타일
+mp_hands = mp.solutions.hands
+
+
+
+# For webcam input:
+cap = cv2.VideoCapture(0)
+width = cap.get(cv2.CAP_PROP_FRAME_WIDTH)
+height = cap.get(cv2.CAP_PROP_FRAME_HEIGHT)
+print('original size: %d, %d' % (width, height))
+
+cap.set(cv2.CAP_PROP_FRAME_WIDTH, width/3)
+cap.set(cv2.CAP_PROP_FRAME_HEIGHT, height/3)
+ 
+width = cap.get(cv2.CAP_PROP_FRAME_WIDTH)
+height = cap.get(cv2.CAP_PROP_FRAME_HEIGHT)
+print('changed size: %d, %d' % (width, height))
+
+
+with mp_hands.Hands(
+    model_complexity=0,
+    max_num_hands=1,
+    min_detection_confidence=0.5, # 손가락 감지 최소 신뢰성(1.0에 가까울수록 신뢰성이 높은 것만 인식)
+    min_tracking_confidence=0.5   # 손가락 추적 최소 신뢰성(1.0에 가까울수록 추적율이 높은 것만 인식)
+    ) as hands:
+  while cap.isOpened():
+    success, image = cap.read()
+    if not success:
+      print("Ignoring empty camera frame.")
+      # If loading a video, use 'break' instead of 'continue'.
+      continue
+
+    # To improve performance, optionally mark the image as not writeable to
+    # pass by reference.
+    image.flags.writeable = False
+    # 사진을 좌/우 반전 시킨후 BGR에서 RGB로 변환
+    #     image = cv2.cvtColor(cv2.flip(image, 1), cv2.COLOR_BGR2RGB)
+
+    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+    # 손 인식 결과
+    results = hands.process(image)
+
+    # Draw the hand annotations on the image.
+    image.flags.writeable = True
+    # 이미지를 다시 RGB에서 BGR로 변경
+    image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+    if results.multi_hand_landmarks:
+      for hand_landmarks in results.multi_hand_landmarks:
+        # ipc 공유메모리를 위한 변수
+        shm = mmap.mmap(0, 512, "Local\\MPipe") 
+        if shm: 
+          for idx in range(0, 21): 
+            shm.write(struct.pack('f', hand_landmarks.landmark[idx].x))
+            shm.write(struct.pack('f', hand_landmarks.landmark[idx].y))
+            shm.write(struct.pack('f', hand_landmarks.landmark[idx].z))        
+        mp_drawing.draw_landmarks(
+            image,
+            hand_landmarks,
+            mp_hands.HAND_CONNECTIONS,
+            mp_drawing_styles.get_default_hand_landmarks_style(),
+            mp_drawing_styles.get_default_hand_connections_style())
+    # Flip the image horizontally for a selfie-view display.
+    cv2.imshow('MediaPipe Hands', cv2.flip(image, 1))
+    if cv2.waitKey(5) & 0xFF == 27:
+      break
+cap.release()
\ No newline at end of file