prompts = args.pop(“prompts“, None) # for SAM-type models

args是一个字典,pop()方法将删除并返回args字典中指定键的值。其中,“prompts”是键名,“None”是键值(如果字典中没有“prompts”这个键名,那么返回None)。所以这行代码的意思是从args字典中取出“prompts”键的值,如果没有则返回None,然后将该键和对应的值从args字典中删除。

  • 1
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
import os import json import csv import cv2 from segment_anything import SamPredictor, sam_model_registry folder_path = 'D:\\segment-anything-main\\segment-anything-main\\input\\Normal\\' # 替换为实际的文件夹路径 output_file = 'D:\\细胞识别\\output.csv' # 替换为实际的输出文件路径 data_list = [] # 用于存储所有的坐标信息 for filename in os.listdir(folder_path): if filename.endswith('.json'): json_path = os.path.join(folder_path, filename) # 读取JSON文件 with open(json_path) as file: data = json.load(file) # 获取多边形坐标 shapes = data['shapes'] polygon_points = shapes[0]['points'] # 假设只有一个多边形标注 # 计算最小包围框的左上角和右下角坐标 x_coordinates = [point[0] for point in polygon_points] y_coordinates = [point[1] for point in polygon_points] min_x = min(x_coordinates) min_y = min(y_coordinates) max_x = max(x_coordinates) max_y = max(y_coordinates) # 将坐标信息添加到列表中 data_list.append({'Filename': filename, 'Min_X': min_x, 'Min_Y': min_y, 'Max_X': max_x, 'Max_Y': max_y}) # 写入CSV文件 with open(output_file, 'w', newline='') as file: fieldnames = ['Filename', 'Min_X', 'Min_Y', 'Max_X', 'Max_Y'] writer = csv.DictWriter(file, fieldnames=fieldnames) writer.writeheader() writer.writerows(data_list) # 生成input_prompts input_prompts = [] for data in data_list: input_prompt = f"处理文件:{data['Filename']},左上角坐标:({data['Min_X']}, {data['Min_Y']}),右下角坐标:({data['Max_X']}, {data['Max_Y']})" input_prompts.append(input_prompt) sam = sam_model_registry["default"](checkpoint="D:\\segment-anything-main\\segment-anything-main\\sam_vit_h_4b8939.pth") predictor = SamPredictor(sam) for filename in os.listdir(folder_path): if filename.lower().endswith(('.png', '.jpg', '.jpeg')): image_path = os.path.join(folder_path, filename) # Load and set the image for prediction your_image = cv2.imread(image_path) predictor.set_image(your_image) # Perform prediction using input prompts masks, _, _ = predictor.predict(input_prompts) # Perform further processing or analysis on the predicted masks for i, mask in enumerate(masks): mask_filename = f"mask_{i + 1}_{filename}" mask_path = os.path.join(folder_path, mask_filename) cv2.imwrite(mask_path, mask)
07-17
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值