手部识别数据集中egohands是比较适合的数据集之一，数据集为mat格式，在网上有些转换的代码是转成csv格式的，还需要后续进行txt的转换，不是太方便，特在原有代码的基础上对数据集标注文件进行txt格式的直接转换。
下载egohands文件后，直接解压，然后在目录内建立一个py文件， 代码如下：
import scipy.io as sioimport osimport gcfrom PIL import Imageimport six.moves.urllib as urllibimport xml.etree.cElementTree as ETimport shutil as shimport numpy as npimport cv2import randomfrom shutil import copyfiledef save_txt(outpath, ans): with open(outpath, 'w') as outfile: outfile.write(ans)def get_bbox_txt(base_path, dir): image_path_array = [] for root, dirs, filenames in os.walk(base_path + dir): for f in filenames: if (f.split(".")[1] == "jpg"): img_path = base_path + dir + "/" + f image_path_array.append(img_path) # sort image_path_array to ensure its in the low to high order expected in polygon.mat image_path_array.sort() pointindex = 0 boxes = sio.loadmat(base_path + dir + "/polygons.mat") # there are 100 of these per folder in the egohands dataset polygons = boxes["polygons"][0] # classLabel for first in polygons: img_id = image_path_array[pointindex] img = cv2.imread(img_id) width = int(np.size(img, 1)) height = int(np.size(img, 0)) pointindex += 1 ans = '' for pointlist in first: # print(np.shape(pst)) #(0,2) max_x = max_y = min_x = min_y = 0 findex = 0 for point in pointlist: if (len(point) == 2): x = int(point[0]) y = int(point[1]) # print('x', x) # print('y', y) if (findex == 0): min_x = x min_y = y findex += 1 max_x = x if (x > max_x) else max_x min_x = x if (x < min_x) else min_x max_y = y if (y > max_y) else max_y min_y = y if (y < min_y) else min_y if (min_x > 0 and min_y > 0 and max_x > 0 and max_y > 0): norm_width = (max_x - min_x) / width norm_height = (max_y - min_y) / height center_x, center_y = (max_x + min_x) / 2, (max_y + min_y) / 2 norm_center_x = center_x / width norm_center_y = center_y / height ans = ans + '2' + ' ' + str(norm_center_x) +' ' +str(norm_center_y) +' ' +str(norm_width) +' ' +str(norm_height) + 'n' txt_path = img_id.split(".")[0] if not os.path.exists(txt_path + ".txt"): save_txt(txt_path + ".txt", ans)def create_directory(dir_path): if not os.path.exists(dir_path): os.makedirs(dir_path)def generate_txt_files(image_dir): for root, dirs, filenames in os.walk(image_dir): for dir in dirs: get_bbox_txt(image_dir, dir) def split_data_test_eval_train(image_dir): create_directory("images") create_directory("images/train") create_directory("images/test") data_size = 4000 loop_index = 0 data_sampsize = int(0.1 * data_size) test_samp_array = random.sample(range(data_size), k=data_sampsize) for root, dirs, filenames in os.walk(image_dir): for dir in dirs: for f in os.listdir(image_dir + dir): if (f.split(".")[1] == "jpg"): loop_index += 1 print(loop_index, f) if loop_index in test_samp_array: os.rename(image_dir + dir + "/" + f, "images/test/" + f) os.rename(image_dir + dir + "/" + f.split(".")[0] + ".txt", "images/test/" + f.split(".")[0] + ".txt") else: os.rename(image_dir + dir + "/" + f, "images/train/" + f) os.rename(image_dir + dir + "/" + f.split(".")[0] + ".txt", "images/train/" + f.split(".")[0] + ".txt") #print(loop_index, image_dir + f) #print("> done scanning director ", dir) os.remove(image_dir + dir + "/polygons.mat") os.rmdir(image_dir + dir)def rename_files(image_dir): print("Renaming files") loop_index = 0 for root, dirs, filenames in os.walk(image_dir): for dir in dirs: for f in os.listdir(image_dir + dir): if (dir not in f): if (f.split(".")[1] == "jpg"): loop_index += 1 os.rename(image_dir + dir + "/" + f, image_dir + dir + "/" + dir + "_" + f) else: break generate_txt_files("_LABELLED_SAMPLES/")rename_files("_LABELLED_SAMPLES/")split_data_test_eval_train("_LABELLED_SAMPLES/")
cd切换到目录内，直接python运行以上代码的py文件，自动在目录内生成一个images文件夹，文件夹内自动匹配了train和val的文件，图片与txt相互对应，非常方便。