点选识别ddddocr
import ddddocr import cv2 from PIL import Image det = ddddocr.DdddOcr(det=True) det2 = ddddocr.DdddOcr(beta=True) with open("download.png", 'rb') as f: image = f.read() bboxes = det.detection(image) # print(bboxes) im = cv2.imread("download.png") all = [] for bbox in bboxes: x1, y1, x2, y2 = bbox # im = cv2.rectangle(im, (x1, y1), (x2, y2), color=(0, 0, 255), thickness=2) # 裁剪出框选区域 cropped = im[y1:y2, x1:x2] # 转换为 RGB 格式(OpenCV 是 BGR) cropped_rgb = cv2.cvtColor(cropped, cv2.COLOR_BGR2RGB) # 转换为 PIL 图像 pil_img = Image.fromarray(cropped_rgb) result = det2.classification(pil_img) tt = f'{result}: {[x1, y1, x2, y2]}' all.append(tt) print(all) # cv2.imwrite("result.jpg", im)