Python .mat图片文件读取
1. .mat文件用h5py读取
.mat格式的NYU v2数据集读取
import numpy as np import h5py import cv2 def h5_loader(path): h5f = h5py.File(path) # .mat文件里有哪些内容 print(h5f) keys = list(h5f.keys()) print("keys: ", keys) # images类别下的数据维度 print("h5f['images'].shape: ", h5f['images'].shape) img = np.array(h5f['images'][0]) print("img.shape: ", img.shape) img = np.transpose(img, (2, 1, 0)) print("img.shape: ", img.shape) depth = np.array(h5f['depths'][0]) print("depth.shape: :", depth.shape) return img, depth if __name__ == '__main__': PATH = 'I:/datasets/NYU Depth Dataset v2/nyu_depth_v2_labeled.mat' img, depth = h5_loader(PATH) img = img.astype('float32') # 保存文件 cv2.imwrite('1.png', img.astype(np.uint8))
结果展示:
<HDF5 file "nyu_depth_v2_labeled.mat" (mode r)> keys: ['#refs#', '#subsystem#', 'accelData', 'depths', 'images', 'instances', 'labels', 'names', 'namesToIds', 'rawDepthFilenames', 'rawDepths', 'rawRgbFilenames', 'sceneTypes', 'scenes'] h5f['images'].shape: (1449, 3, 640, 480) img.shape: (3, 640, 480) img.shape: (480, 640, 3) depth.shape: : (640, 480)
图片:
2. 结语
努力去爱周围的每一个人,付出,不一定有收获,但是不付出就一定没有收获! 给街头卖艺的人零钱,不和深夜还在摆摊的小贩讨价还价。愿我的博客对你有所帮助(*^▽^*)(*^▽^*)!
如果客官喜欢小生的园子,记得关注小生哟,小生会持续更新(#^.^#)(#^.^#)。
但行好事 莫问前程