NNES (Non-connected negative elements scanning) can identify the coarse background. (a) Performance of a brightness-based adaptive global thresholding using peak-of-frequency of brightness as a feature value. A random set of 31 actin image of 25*800*800 in our database are used to test the performance. Left, the frequency distribution curves of the pixel brightness of the 31 training samples; right, the linear correlation of MGT (manual global threshold)-determined coarse background (as ground truth) vs. the threshold at peak-of-frequency. This method cannot identify the peak accurately due to the turbulent curve and the results do not have good correlation. (b) NNE count of pure ground noise has a normal-like distribution. A ground noise image was mimicked by a random normal distribution (mean = 90, std = 30) into a 25*800*800 array. The maximum projection is conducted by choosing the maximum value of the third axis to make an 800*800 image (distribution shown as blue), and its mean and STD are calculated to make a true normal distribution (shown as red), both of which finely overlap. (c) Performance of the NNES-based adaptive global thresholding and its prediction. The same samples of (a) is used for the assay. Left, the NNES curves reflecting the relationship between the threshold and NNE count; right, correlation of ground truth coarse background evaluated by MGT vs. peak of NNE count for individual samples. Comparing to the brightness-based methods, NNES has a much smoother shape that enables the utilization of peak as a feature value. Also, NNES has a more robust correlation to coarse background determined by MGT.