PROGRAMMING

1. track.py from tracking import Tracker, Trackable import cv2 import numpy as np import time frame_size = 416 frame_count = 0 min_confidence = 0.5 min_directions = 10 height = 0 width = 0 count_limit = 0 up_count = 0 down_count = 0 direction = '' trackers = [] trackables = {} file_name = 'Object_tracking/Testvideo.mp4' output_name = 'Object_tracking/yolo_passenger_counting/Testvideo_out_01.avi'..

import cv2 import numpy as np bins = np.arange(256).reshape(256, 1) def draw_histogram(img): h = np.zeros((img.shape[0], 256), dtype = np.uint8) hist_item = cv2.calcHist([img], [0], None, [256], [0, 256]) cv2.normalize(hist_item, hist_item, 0, 255, cv2.NORM_MINMAX) hist = np.int32(np.around(hist_item)) for x, y in enumerate(hist): cv2.line(h, (x, 10), (x, int(y) + 10), (255, 255, 255)) cv2.line(..

In [ ]: #pip install selectivesearch In [2]: import selectivesearch import cv2 import matplotlib.pyplot as plt %matplotlib inline In [3]: img = cv2.imread('circle.jpg', cv2.IMREAD_ANYCOLOR) img_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) plt.figure(figsize = (8, 8)) plt.imshow(img_rgb) plt.show() In [4]: _, regions = selectivesearch.selective_search(img_rgb, scale = 100, min_size = 2000) ..

1. 3인 가위바위보 - TIE 케이스 2. 3인 가위바위보 - Win / Lose 케이스

In [ ]: 1. GPU Setup¶ 메뉴에서 '런타임' > '런타임 유형 변경' > 노트 설정 > 하드웨어 가속기 > 'GPU' 선택 2. Mount Google Drive¶ In [1]: from google.colab import drive drive.mount('/content/drive') Mounted at /content/drive In [ ]: !ls drive sample_data 3. Install CUDA related modules, cuDNN¶ Download cuDNN file https://developer.nvidia.com/rdp/cudnn-download Unzip the cuDNN from Google Drive to Colab /usr/local/cuda folder..

In [1]: from google.colab import drive drive.mount('/content/drive') Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True). In [2]: !ls -la "/content/drive/MyDrive/darknet" total 32 drwx------ 2 root root 4096 Dec 23 03:16 backup drwx------ 2 root root 4096 Dec 23 02:33 bin drwx------ 2 root root 4096 Dec 23 03:04 cfg drwx..

1. 웹카메라 연결 2. yolo_custom_test_video.py import numpy as np import cv2 import time min_confidence = 0.5 width = 1280 height = 0 show_ratio = 1.0 title_name = 'Custom Yolo' # Load Yolo net = cv2.dnn.readNet("./machineVision/yolo_custom_test_video/model/custom-train-yolo_final.weights", "./machineVision/yolo_custom_test_video/custom/custom-train-yolo.cfg") classes = [] with open("./machineVision/yo..

1. 2. 3. 4. 5. 6. 7. 8. 이어서 학습하기 0xlordfo.tistory.com/9 YOLO 학습에 대한 잡다한 지식(기본, 배경 지식) 1. windows 버전 darknet이 darknet_no_gpu.sln 버전과 darknet.sln으로 나뉜 이유. YOLO는 joseph redmon이라는 사람이 만들어 냈다. github나 트위터 등에서는 pjreddie라는 닉네임으로 활동하고 있다. YOLO는.. 0xlordfo.tistory.com