def read_mrc_data(path, show_progress=False): #return read_mrc(path=path, show_progress=show_progress)['value'] return TIM.read_data(path)
def read_mrc_data(path, show_progress=False): return TIM.read_data(path)
@jit(nopython=True) def generate_feature_vector(b, label, cluster_center_number): result = np.array([0] * cluster_center_number) # sum_f = np.array((cluster_center_number, 2), 0) sum_f = np.array([[0 for i in range(2)] for j in range(cluster_center_number)]) for i in range(0, b.shape[0]): for j in range(0, b.shape[1]): for k in range(0, b.shape[2]): sum_f[label[i][j][k]][1] = sum_f[label[i][j][k]][1] + 1 sum_f[label[i][j][k]][0] = sum_f[label[i][j][k]][0] + b[i][j][k] for i in range(cluster_center_number): assert sum_f[i][1] > 0 result[i] = sum_f[i][0] / sum_f[i][1] return result if __name__ == "__main__": # file path path = input("Enter data path: ") from aitom.io import mrcfile_proxy a = mrcfile_proxy.read_data(path) print("file has been read, shape is", a.shape) start_time = time.time() saliency_detection(a=a, gaussian_sigma=2.5, gabor_sigma=14.0, gabor_lambda=13.0, cluster_center_number=10000, multiprocessing_num=0, pick_num=1000, save_flag=True) end_time = time.time() print('saliency detection takes', end_time - start_time, 's')
dmax = np.amax(dimg) dimg = (dimg / dmax) * 255 # TODO: Change the image saving path img_path = '/Users/apple/Desktop/Lab/Zach_Project/Denoising_Result/Difference_' + str( type) + '.png' plt.imsave(img_path, dimg, cmap='gray') if __name__ == "__main__": # TODO: Change the data path, name and Gussian denoise type path = '/Users/apple/Desktop/Lab/Zach_Project/Sample_Data/aitom_demo_single_particle_tomogram.mrc' name = 'aitom_demo_single_particle_tomogram' G_type = 1 # read the volume data as numpy array original = mrcfile_proxy.read_data(path) # save a slice of original tomogram oimg = (original[:, :, int(original.shape[2] / 2)]).copy() # TODO: Change the orginal image saving directory img_path = '/Users/apple/Desktop/Lab/Zach_Project/Denoising_Result/Original.png' plt.imsave(img_path, oimg, cmap='gray') # perform three diffrent kind of denoising # The difference comparison between original tomogram and filtered tomogram is commented out g_fimg = g_denoising(G_type, a=original, name=name, gaussian_sigma=2.5, save_flag=True) # diff_compare(oimg, g_fimg, "Gaussian")