以下是删除了所有注释后的代码： ```python import matplotlib.pyplot as plt import numpy as np import os import cv2 import tifffile as tiff import datetime import time class RCNCalctor(object): imgNum = 0 imgWidth = 0 imgHeight = 0 src_path = '' cutoffx = 0 binx = 1 images = [] channelNumPerChip = 256 roiWidth = channelNumPerChip roiHeight = channelNumPerChip numOfColRoi = imgWidth / channelNumPerChip def __init__(self, *args, **kwargs): return super().__init__(*args, **kwargs) def load(self, src_path, binx, cutoffx): self.src_path = src_path self.binx = binx self.cutoffx = cutoffx self.images = tiff.imread(src_path) self.imgNum = self.images.shape[0] self.imgWidth = self.images.shape[2] self.imgHeight = self.images.shape[1] self.roiWidth = self.roiWidth // binx self.roiHeight = self.roiHeight // binx self.numOfColRoi = int(self.imgWidth / self.channelNumPerChip * binx) print(self.images.shape) print('read image successfully!') def testNPDimension(self, src_path): img0 = tiff.imread(src_path) print(img0.shape) roiImg = img0[0:32, 1:65] print(roiImg.shape) def calc(self): if self.imgNum <= 1: raise AssertionError("image number should larger than 1!") self.numOfColRoi = int(self.imgWidth / self.roiWidth) darkAvgs = np.zeros((self.imgNum, 1), dtype=np.float64) avgImg = np.zeros((self.imgHeight, self.imgWidth), dtype=np.float64) stdImg = np.zeros((self.imgHeight, self.imgWidth), dtype=np.float64) all_std=np.zeros((1,self.imgWidth),dtype=np.float64) for i in range(self.imgWidth): all_std=all_std+np.array(np.sqrt(np.mean(self.images[:, :,i ] - self.images[:, :, i].mean()) ** 2),dtype=np.float64) for idx in range(self.imgNum): darkAvgs[idx, 0] = cv2.mean(self.images[idx, :, :])[0] avgImg = avgImg + np.array(self.images[idx, :, :], dtype=np.float64) stdImg = stdImg + np.array(self.images[idx, :, :], dtype=np.float64) * np.array(self.images[idx, :, :], dtype=np.float64) avgImg = avgImg / self.imgNum stdImg = (stdImg - avgImg * avgImg * self.imgNum) / (self.imgNum - 1) stdImg = np.sqrt(stdImg) stdAvg = cv2.mean(stdImg)[0] stdAvgPerChip = np.zeros((self.numOfColRoi, 1), dtype=np.float64) for i in range(self.numOfColRoi): roiImg = stdImg[:, i * self.roiWidth:(i + 1) * self.roiWidth] stdAvgPerChip[i] = cv2.mean(roiImg)[0] minValue, maxValue, minLoc, maxLoc = cv2.minMaxLoc(darkAvgs) meanValue = np.mean(darkAvgs, 0) stability = abs(maxValue - minValue) / meanValue print('for image sequences {0}, binSize: {1}, cutoff:{2}'.format(self.src_path, self.binx, self.cutoffx)) print('stability:%.2f%%' % (stability * 100)) print('average random noise: %.2f' % (stdAvg)) strStdAvgPerChips = ['%.2f' % (value) for value in stdAvgPerChip] print('random noise per chip:') print(' '.join(strStdAvgPerChips)) offY = [] if self.roiHeight > self.imgHeight or self.roiWidth > self.imgWidth: raise Exception('roi size ({0}x{1}) is larger than image size ({0}x{1})'.format(self.roiWidth, self.roiHeight, self.imgWidth, self.imgHeight)) if self.roiHeight < self.imgHeight / 2: offY = [int(self.imgHeight / 4 - self.roiHeight / 2), int(self.imgHeight * 3 / 4 - self.roiHeight / 2)] else: offY = [int(self.imgHeight - self.roiHeight) / 2] RCN2D = np.zeros((self.imgNum, len(offY), self.numOfColRoi), dtype=np.float64) RN2D = np.zeros((self.imgNum, len(offY), self.numOfColRoi), dtype=np.float64) CN2D = np.zeros((self.imgNum, len(offY), self.numOfColRoi), dtype=np.float64) for idx in range(self.imgNum): noiseImg = np.array(self.images[idx, :, :], dtype=np.float64) - avgImg noiseImg[0:self.cutoffx, :] = noiseImg[self.cutoffx:2 * self.cutoffx, :] noiseImg[(self.imgHeight - self.cutoffx):self.imgHeight, :] = noiseImg[(self.imgHeight - 2 * self.cutoffx):(self.imgHeight - self.cutoffx), :] noiseImg[:, 0:self.cutoffx] = noiseImg[:, self.cutoffx:2 * self.cutoffx] noiseImg[:, (self.imgWidth - self.cutoffx):self.imgWidth] = noiseImg[:, (self.imgWidth - 2 * self.cutoffx):(self.imgWidth - self.cutoffx)] for h in range(len(offY)): for w in range(self.numOfColRoi): roiImg = noiseImg[int(offY[h]):int(offY[h] + self.roiHeight), int(w * self.roiWidth):int(w * self.roiWidth + self.roiWidth)] rowAvgs = np.mean(roiImg, 1) rmed = np.median(rowAvgs) tmpRNInRow = np.std(rowAvgs) rowIdxs = [i for i in range(len(rowAvgs)) if abs(rowAvgs[i] - rmed) >= 3.5 * tmpRNInRow] rowAvgs[rowIdxs] = rmed RNInRow = np.std(rowAvgs) colAvgs = np.mean(roiImg, 0) cmed = np.median(colAvgs) tmpCNInCol = np.std(colAvgs) colIdxs = [i for i in range(len(colAvgs)) if abs(colAvgs[i] - cmed) >= 3.5 * tmpCNInCol] colAvgs[colIdxs] = cmed RNInCol = np.std(colAvgs) RN2D[idx][h][w] = RNInRow CN2D[idx][h][w] = RNInCol RCN2D[idx][h][w] = RNInRow / max(RNInCol, 1e-6) avgRCN = np.mean(RCN2D, 0) avgRN = np.mean(RN2D, 0) avgCN = np.mean(CN2D, 0) chipAvgRCN = np.mean(avgRCN, 0) chipAvgRN = np.mean(avgRN, 0) chipAvgCN = np.mean(avgCN, 0) medRCN = np.median(RCN2D, 0) medRN = np.median(RN2D, 0) medCN = np.median(CN2D, 0) chipMedRCN = np.mean(medRCN, 0) chipMedRN = np.mean(medRN, 0) chipMedCN = np.mean(medCN, 0) logDir = 'logs' if not os.path.exists(logDir): os.makedirs(logDir) dt = datetime.datetime.now() strTime = dt.strftime('%Y-%m-%d') fid = open(os.path.join(logDir, strTime + '.log'), 'a') fid.write('for image sequences {0}, binSize: {1}, cutoff:{2}\n'.format(self.src_path, self.binx, self.cutoffx)) fid.write('stablility: %.2f%%\n' % (stability * 100)) fid.write('average random noise: %.2f\n' % (stdAvg)) fid.write('random noise per chip:\n') fid.write(' '.join(strStdAvgPerChips) + '\n') fid.write('average of row noise per chip:\n') strAvgRNs = (['%.2f' % (value) for value in chipAvgRCN]) strAvgRN = ' '.join(strAvgRNs) fid.write(strAvgRN + '\n') fid.write('average of column noise per chip:\n') strAvgCN = ['%.2f' % (value) for value in chipAvgCN] fid.write(' '.join(strAvgCN) + '\n') fid.write('median of row noise per chip:\n') strMedRN = ['%.2f' % (value) for value in chipMedRN] fid.write(' '.join(strMedRN) + '\n') fid.write('median of column noise per chip:\n') strMedCN = ['%.2f' % (value) for value in chipMedCN] fid.write(' '.join(strMedCN) + '\n') fid.write('average of RCN per chip:\n') strAvgRCN = ['%.2f' % (value) for value in chipAvgRCN] fid.write(' '.join(strAvgRCN) + '\n') fid.write('median of RCN per chip:\n') strMedRCN = ['%.2f' % (value) for value in chipMedRCN] fid.write(' '.join(strMedRCN) + '\n') fid.write('\n') fid.close() print('median of RCN per chip:') print(' '.join(strMedRCN)) return stability, stdAvg, chipMedRCN, chipAvgRCN if __name__ == "__main__": calctor = RCNCalctor() calctor.load(f"G:\\xr\\devtools\\RCN\\TestCodes\\refer\\F2332 10fps Gain4 暗场图.tif", 4, 4) calctor.calc() ``` 请注意，这段代码已经去除了所有的注释行。如果你有进一步的需求或需要对代码进行特定修改，请告知！