{"id":979130,"date":"2024-12-27T06:43:49","date_gmt":"2024-12-26T22:43:49","guid":{"rendered":""},"modified":"2024-12-27T06:43:51","modified_gmt":"2024-12-26T22:43:51","slug":"%e5%a6%82%e4%bd%95%e7%94%a8python%e8%af%86%e5%88%ab%e5%9b%be%e7%89%87","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/979130.html","title":{"rendered":"\u5982\u4f55\u7528python\u8bc6\u522b\u56fe\u7247"},"content":{"rendered":"

\"\u5982\u4f55\u7528python\u8bc6\u522b\u56fe\u7247\"<\/p>\n

\u4e00\u3001\u5982\u4f55\u7528Python\u8bc6\u522b\u56fe\u7247<\/p>\n<\/p>\n

Python\u8bc6\u522b\u56fe\u7247\u7684\u65b9\u6cd5\u4e3b\u8981\u5305\u62ec\uff1a\u4f7f\u7528OpenCV\u5e93\u8fdb\u884c\u56fe\u50cf\u5904\u7406\u3001\u5229\u7528PIL\u5e93\u8fdb\u884c\u57fa\u672c\u56fe\u50cf\u64cd\u4f5c\u3001\u901a\u8fc7TensorFlow\u6216PyTorch\u8fdb\u884c\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\u9884\u6d4b\u3002<\/strong>\u5728\u8fd9\u4e9b\u65b9\u6cd5\u4e2d\uff0c\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\u901a\u5e38\u80fd\u591f\u63d0\u4f9b\u66f4\u52a0\u51c6\u786e\u548c\u590d\u6742\u7684\u56fe\u50cf\u8bc6\u522b\u529f\u80fd\u3002\u6bd4\u5982\uff0c\u4f7f\u7528\u9884\u8bad\u7ec3\u7684\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\uff08CNN\uff09\u6a21\u578b\u53ef\u4ee5\u5feb\u901f\u8bc6\u522b\u56fe\u7247\u4e2d\u7684\u7269\u4f53\u7c7b\u522b\u3002\u63a5\u4e0b\u6765\uff0c\u6211\u4eec\u5c06\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528\u8fd9\u4e9b\u65b9\u6cd5\u6765\u5b9e\u73b0\u56fe\u7247\u8bc6\u522b\u3002<\/p>\n<\/p>\n

\u4e8c\u3001OpenCV\u8fdb\u884c\u56fe\u50cf\u5904\u7406<\/p>\n<\/p>\n

OpenCV\u662f\u4e00\u4e2a\u529f\u80fd\u5f3a\u5927\u7684\u8ba1\u7b97\u673a\u89c6\u89c9\u5e93\uff0c\u652f\u6301\u591a\u79cd\u7f16\u7a0b\u8bed\u8a00\uff0c\u5305\u62ecPython\u3002\u5b83\u63d0\u4f9b\u4e86\u4e00\u7cfb\u5217\u7684\u56fe\u50cf\u5904\u7406\u529f\u80fd\uff0c\u53ef\u4ee5\u7528\u6765\u5b9e\u73b0\u56fe\u50cf\u8bc6\u522b\u3002<\/p>\n<\/p>\n

    \n
  1. \u5b89\u88c5OpenCV<\/strong><\/li>\n<\/ol>\n

    \u9996\u5148\uff0c\u786e\u4fdd\u4f60\u7684Python\u73af\u5883\u4e2d\u5b89\u88c5\u4e86OpenCV\u5e93\uff0c\u53ef\u4ee5\u901a\u8fc7pip\u5b89\u88c5\uff1a<\/p>\n<\/p>\n

    pip install opencv-python<\/p>\n
    <\/code><\/pre>\n<\/p>\n
      \n
    1. \u8bfb\u53d6\u548c\u663e\u793a\u56fe\u50cf<\/strong><\/li>\n<\/ol>\n
      \u4f7f\u7528OpenCV\u53ef\u4ee5\u8f7b\u677e\u5730\u8bfb\u53d6\u548c\u663e\u793a\u56fe\u50cf\uff1a<\/p>\n<\/p>\n
      import cv2<\/p>\n
      \u8bfb\u53d6\u56fe\u50cf<\/strong><\/h2>\n
      image = cv2.imread('path_to_image.jpg')<\/p>\n
      \u663e\u793a\u56fe\u50cf<\/strong><\/h2>\n
      cv2.imshow('Image', image)<\/p>\n
      cv2.wAI<\/a>tKey(0)<\/p>\n
      cv2.destroyAllWindows()<\/p>\n
      <\/code><\/pre>\n<\/p>\n
        \n
      1. \u56fe\u50cf\u9884\u5904\u7406<\/strong><\/li>\n<\/ol>\n
        \u56fe\u50cf\u8bc6\u522b\u901a\u5e38\u9700\u8981\u4e00\u4e9b\u9884\u5904\u7406\u6b65\u9aa4\uff0c\u6bd4\u5982\u7070\u5ea6\u5316\u3001\u8fb9\u7f18\u68c0\u6d4b\u7b49\uff1a<\/p>\n<\/p>\n
        # \u8f6c\u6362\u4e3a\u7070\u5ea6\u56fe\u50cf<\/p>\n
        gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n
        \u4f7f\u7528Canny\u8fb9\u7f18\u68c0\u6d4b<\/strong><\/h2>\n
        edges = cv2.Canny(gray_image, 100, 200)<\/p>\n
        <\/code><\/pre>\n<\/p>\n
          \n
        1. \u7279\u5f81\u68c0\u6d4b<\/strong><\/li>\n<\/ol>\n
          OpenCV\u63d0\u4f9b\u4e86\u591a\u79cd\u7279\u5f81\u68c0\u6d4b\u7b97\u6cd5\uff0c\u6bd4\u5982SIFT\u3001SURF\u7b49\uff0c\u53ef\u4ee5\u7528\u6765\u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u5173\u952e\u7279\u5f81\uff1a<\/p>\n<\/p>\n
          # \u521b\u5efaSIFT\u7279\u5f81\u68c0\u6d4b\u5668<\/p>\n
          sift = cv2.SIFT_create()<\/p>\n
          \u68c0\u6d4b\u5173\u952e\u70b9<\/strong><\/h2>\n
          keypoints, descriptors = sift.detectAndCompute(gray_image, None)<\/p>\n
          \u7ed8\u5236\u5173\u952e\u70b9<\/strong><\/h2>\n
          image_with_keypoints = cv2.drawKeypoints(image, keypoints, None)<\/p>\n
          cv2.imshow('SIFT Keypoints', image_with_keypoints)<\/p>\n
          cv2.waitKey(0)<\/p>\n
          cv2.destroyAllWindows()<\/p>\n
          <\/code><\/pre>\n<\/p>\n
          \u4e09\u3001PIL\u5e93\u8fdb\u884c\u57fa\u672c\u56fe\u50cf\u64cd\u4f5c<\/p>\n<\/p>\n
          PIL\uff08Pillow\uff09\u662f\u4e00\u4e2a\u7b80\u5355\u6613\u7528\u7684\u56fe\u50cf\u5904\u7406\u5e93\uff0c\u53ef\u4ee5\u5904\u7406\u56fe\u50cf\u7684\u57fa\u672c\u64cd\u4f5c\u3002<\/p>\n<\/p>\n
            \n
          1. \u5b89\u88c5Pillow<\/strong><\/li>\n<\/ol>\n
            \u9996\u5148\u5b89\u88c5Pillow\u5e93\uff1a<\/p>\n<\/p>\n
            pip install pillow<\/p>\n
            <\/code><\/pre>\n<\/p>\n
              \n
            1. \u8bfb\u53d6\u548c\u663e\u793a\u56fe\u50cf<\/strong><\/li>\n<\/ol>\n
              \u4f7f\u7528Pillow\u53ef\u4ee5\u8bfb\u53d6\u548c\u663e\u793a\u56fe\u50cf\uff1a<\/p>\n<\/p>\n
              from PIL import Image<\/p>\n
              \u8bfb\u53d6\u56fe\u50cf<\/strong><\/h2>\n
              image = Image.open('path_to_image.jpg')<\/p>\n
              \u663e\u793a\u56fe\u50cf<\/strong><\/h2>\n
              image.show()<\/p>\n
              <\/code><\/pre>\n<\/p>\n
                \n
              1. \u56fe\u50cf\u8f6c\u6362<\/strong><\/li>\n<\/ol>\n
                \u53ef\u4ee5\u5bf9\u56fe\u50cf\u8fdb\u884c\u5404\u79cd\u8f6c\u6362\uff0c\u6bd4\u5982\u8c03\u6574\u5927\u5c0f\u3001\u65cb\u8f6c\u3001\u526a\u88c1\u7b49\uff1a<\/p>\n<\/p>\n
                # \u8c03\u6574\u5927\u5c0f<\/p>\n
                resized_image = image.resize((100, 100))<\/p>\n
                \u65cb\u8f6c\u56fe\u50cf<\/strong><\/h2>\n
                rotated_image = image.rotate(45)<\/p>\n
                \u526a\u88c1\u56fe\u50cf<\/strong><\/h2>\n
                cropped_image = image.crop((10, 10, 100, 100))<\/p>\n
                <\/code><\/pre>\n<\/p>\n
                \u56db\u3001\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\u8fdb\u884c\u56fe\u50cf\u8bc6\u522b<\/p>\n<\/p>\n
                \u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u5c24\u5176\u662f\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\uff08CNN\uff09\uff0c\u5728\u56fe\u50cf\u8bc6\u522b\u9886\u57df\u8868\u73b0\u51fa\u8272\u3002\u4e0b\u9762\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528TensorFlow\/Keras\u8fdb\u884c\u56fe\u50cf\u8bc6\u522b\u3002<\/p>\n<\/p>\n
                  \n
                1. \u5b89\u88c5TensorFlow<\/strong><\/li>\n<\/ol>\n
                  \u9996\u5148\uff0c\u5b89\u88c5TensorFlow\u5e93\uff1a<\/p>\n<\/p>\n
                  pip install tensorflow<\/p>\n
                  <\/code><\/pre>\n<\/p>\n
                    \n
                  1. \u52a0\u8f7d\u9884\u8bad\u7ec3\u6a21\u578b<\/strong><\/li>\n<\/ol>\n
                    TensorFlow\u63d0\u4f9b\u4e86\u4e00\u4e9b\u9884\u8bad\u7ec3\u7684\u6a21\u578b\uff0c\u53ef\u4ee5\u7528\u6765\u8fdb\u884c\u56fe\u50cf\u5206\u7c7b\uff1a<\/p>\n<\/p>\n
                    import tensorflow as tf<\/p>\n
                    from tensorflow.keras.applications import ResNet50<\/p>\n
                    from tensorflow.keras.preprocessing import image<\/p>\n
                    from tensorflow.keras.applications.resnet50 import preprocess_input, decode_predictions<\/p>\n
                    import numpy as np<\/p>\n
                    \u52a0\u8f7dResNet50\u6a21\u578b<\/strong><\/h2>\n
                    model = ResNet50(weights='imagenet')<\/p>\n
                    \u8bfb\u53d6\u548c\u9884\u5904\u7406\u56fe\u50cf<\/strong><\/h2>\n
                    img_path = 'path_to_image.jpg'<\/p>\n
                    img = image.load_img(img_path, target_size=(224, 224))<\/p>\n
                    x = image.img_to_array(img)<\/p>\n
                    x = np.expand_dims(x, axis=0)<\/p>\n
                    x = preprocess_input(x)<\/p>\n
                    \u9884\u6d4b<\/strong><\/h2>\n
                    preds = model.predict(x)<\/p>\n
                    \u6253\u5370\u9884\u6d4b\u7ed3\u679c<\/strong><\/h2>\n
                    print('Predicted:', decode_predictions(preds, top=3)[0])<\/p>\n
                    <\/code><\/pre>\n<\/p>\n
                      \n
                    1. \u81ea\u5b9a\u4e49\u6a21\u578b\u8bad\u7ec3<\/strong><\/li>\n<\/ol>\n
                      \u5982\u679c\u9700\u8981\u8fdb\u884c\u66f4\u4e3a\u590d\u6742\u7684\u56fe\u50cf\u8bc6\u522b\u4efb\u52a1\uff0c\u53ef\u4ee5\u4f7f\u7528\u81ea\u5b9a\u4e49\u6570\u636e\u96c6\u8bad\u7ec3\u6a21\u578b\uff1a<\/p>\n<\/p>\n
                      from tensorflow.keras.models import Sequential<\/p>\n
                      from tensorflow.keras.layers import Dense, Conv2D, MaxPooling2D, Flatten<\/p>\n
                      from tensorflow.keras.preprocessing.image import ImageDataGenerator<\/p>\n
                      \u521b\u5efa\u7b80\u5355\u7684CNN\u6a21\u578b<\/strong><\/h2>\n
                      model = Sequential([<\/p>\n
                          Conv2D(32, (3, 3), activation='relu', input_shape=(64, 64, 3)),<\/p>\n
                          MaxPooling2D(pool_size=(2, 2)),<\/p>\n
                          Flatten(),<\/p>\n
                          Dense(128, activation='relu'),<\/p>\n
                          Dense(1, activation='sigmoid')<\/p>\n
                      ])<\/p>\n
                      \u7f16\u8bd1\u6a21\u578b<\/strong><\/h2>\n
                      model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])<\/p>\n
                      \u4f7f\u7528ImageDataGenerator\u8fdb\u884c\u6570\u636e\u589e\u5f3a<\/strong><\/h2>\n
                      train_datagen = ImageDataGenerator(rescale=1.\/255)<\/p>\n
                      train_generator = train_datagen.flow_from_directory(<\/p>\n
                          'path_to_training_data',<\/p>\n
                          target_size=(64, 64),<\/p>\n
                          batch_size=32,<\/p>\n
                          class_mode='binary')<\/p>\n
                      \u8bad\u7ec3\u6a21\u578b<\/strong><\/h2>\n
                      model.fit(train_generator, epochs=10)<\/p>\n
                      <\/code><\/pre>\n<\/p>\n
                      \u4e94\u3001\u603b\u7ed3<\/p>\n<\/p>\n
                      \u5728Python\u4e2d\u5b9e\u73b0\u56fe\u7247\u8bc6\u522b\u53ef\u4ee5\u91c7\u7528\u591a\u79cd\u65b9\u6cd5\uff0c\u5176\u4e2d\u5305\u62ec\u4f7f\u7528OpenCV\u8fdb\u884c\u56fe\u50cf\u5904\u7406\u3001\u5229\u7528PIL\u8fdb\u884c\u57fa\u7840\u56fe\u50cf\u64cd\u4f5c\uff0c\u4ee5\u53ca\u901a\u8fc7\u6df1\u5ea6\u5b66\u4e60\u6846\u67b6\u5982TensorFlow\u8fdb\u884c\u66f4\u9ad8\u7ea7\u7684\u56fe\u50cf\u5206\u7c7b\u4efb\u52a1\u3002\u9009\u62e9\u5408\u9002\u7684\u65b9\u6cd5\u53d6\u51b3\u4e8e\u5177\u4f53\u7684\u5e94\u7528\u573a\u666f\u548c\u8bc6\u522b\u4efb\u52a1\u7684\u590d\u6742\u6027<\/strong>\u3002OpenCV\u9002\u5408\u5904\u7406\u57fa\u672c\u7684\u56fe\u50cf\u5904\u7406\u4efb\u52a1\uff0c\u800cTensorFlow\u548cPyTorch\u7b49\u6df1\u5ea6\u5b66\u4e60\u6846\u67b6\u5219\u53ef\u4ee5\u5e94\u5bf9\u66f4\u4e3a\u590d\u6742\u7684\u56fe\u50cf\u8bc6\u522b\u6311\u6218\u3002\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\uff0c\u6839\u636e\u9700\u8981\u9009\u62e9\u5408\u9002\u7684\u5de5\u5177\u548c\u6280\u672f\uff0c\u4ee5\u5b9e\u73b0\u6700\u4f73\u7684\u56fe\u50cf\u8bc6\u522b\u6548\u679c\u3002<\/p>\n<\/p>\n
                      \u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n
                       Python\u53ef\u4ee5\u8bc6\u522b\u54ea\u4e9b\u7c7b\u578b\u7684\u56fe\u7247\uff1f<\/strong>
                      Python\u53ef\u4ee5\u901a\u8fc7\u5404\u79cd\u5e93\u6765\u8bc6\u522b\u548c\u5904\u7406\u591a\u79cd\u7c7b\u578b\u7684\u56fe\u7247\uff0c\u5305\u62ecJPEG\u3001PNG\u3001GIF\u3001BMP\u7b49\u5e38\u89c1\u683c\u5f0f\u3002\u4f7f\u7528OpenCV\u3001Pillow\u6216TensorFlow\u7b49\u5e93\uff0c\u53ef\u4ee5\u5bf9\u56fe\u50cf\u8fdb\u884c\u5206\u7c7b\u3001\u76ee\u6807\u68c0\u6d4b\u3001\u8fb9\u7f18\u68c0\u6d4b\u7b49\u64cd\u4f5c\uff0c\u6ee1\u8db3\u4e0d\u540c\u5e94\u7528\u573a\u666f\u7684\u9700\u6c42\u3002<\/p>\n
                      \u5728Python\u4e2d\u8fdb\u884c\u56fe\u7247\u8bc6\u522b\u9700\u8981\u54ea\u4e9b\u5e93\uff1f<\/strong>
                      \u8fdb\u884c\u56fe\u7247\u8bc6\u522b\u7684\u5e38\u7528\u5e93\u5305\u62ecOpenCV\u3001Pillow\u3001TensorFlow\u3001Keras\u548cPyTorch\u7b49\u3002OpenCV\u9002\u5408\u8fdb\u884c\u57fa\u7840\u7684\u56fe\u50cf\u5904\u7406\u548c\u8ba1\u7b97\u673a\u89c6\u89c9\u4efb\u52a1\uff0cPillow\u5219\u7528\u4e8e\u7b80\u5355\u7684\u56fe\u50cf\u64cd\u4f5c\uff0c\u800cTensorFlow\u548cKeras\u5219\u9002\u5408\u6784\u5efa\u548c\u8bad\u7ec3\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u7528\u4e8e\u66f4\u590d\u6742\u7684\u56fe\u50cf\u8bc6\u522b\u4efb\u52a1\u3002<\/p>\n
                      \u5982\u4f55\u63d0\u9ad8Python\u56fe\u7247\u8bc6\u522b\u7684\u51c6\u786e\u6027\uff1f<\/strong>
                      \u63d0\u5347\u56fe\u7247\u8bc6\u522b\u51c6\u786e\u6027\u7684\u65b9\u6cd5\u6709\u5f88\u591a\u3002\u9996\u5148\uff0c\u53ef\u4ee5\u4f7f\u7528\u66f4\u9ad8\u8d28\u91cf\u7684\u8bad\u7ec3\u6570\u636e\u96c6\uff0c\u786e\u4fdd\u6570\u636e\u7684\u591a\u6837\u6027\u548c\u4ee3\u8868\u6027\u3002\u5176\u6b21\uff0c\u5c1d\u8bd5\u4e0d\u540c\u7684\u6a21\u578b\u67b6\u6784\u548c\u8d85\u53c2\u6570\u4f18\u5316\uff0c\u5229\u7528\u6570\u636e\u589e\u5f3a\u6280\u672f\u6765\u589e\u52a0\u6a21\u578b\u7684\u9c81\u68d2\u6027\u3002\u6b64\u5916\uff0c\u8fc1\u79fb\u5b66\u4e60\u4e5f\u662f\u4e00\u79cd\u6709\u6548\u7684\u65b9\u5f0f\uff0c\u53ef\u4ee5\u5229\u7528\u9884\u8bad\u7ec3\u6a21\u578b\u6765\u63d0\u9ad8\u8bc6\u522b\u6027\u80fd\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"\u4e00\u3001\u5982\u4f55\u7528Python\u8bc6\u522b\u56fe\u7247 Python\u8bc6\u522b\u56fe\u7247\u7684\u65b9\u6cd5\u4e3b\u8981\u5305\u62ec\uff1a\u4f7f\u7528OpenCV\u5e93\u8fdb\u884c\u56fe\u50cf\u5904\u7406\u3001\u5229\u7528PIL […]","protected":false},"author":3,"featured_media":979135,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[37],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/979130"}],"collection":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/comments?post=979130"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/979130\/revisions"}],"predecessor-version":[{"id":979136,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/979130\/revisions\/979136"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/979135"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=979130"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=979130"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=979130"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}