The neuropil, the plexus of axons and dendrites, plays a critical role in operating the circuit processing of the nervous system. Revealing the spatiotemporal activity pattern within the neuropil would clarify how the information flows throughout the nervous system. However, calcium imaging to examine the circuit dynamics has mainly focused on the population of somas due to their discrete distribution. Development of a methodology to analyze the calcium imaging data of the densely packed neuropil would provide us with new insights into the circuit dynamics.Here, we propose a new method to decompose calcium imaging data of the neuropil into populations of bouton-like synaptic structures. To this aim, we pan-neuronally expressed a membrane-bound form of the calcium probe protein, CD4::GCaMP6f in Drosophila larvae. With the probe, we succeeded in visualizing a calcium signal in the neuropil. We obtained the baseline of the signal in the spatial- and temporal-dimension by an iteration method, which enabled the alignment of the neuropil at a single bouton scale and attenuation of the sample deformation and stage drift during the imaging. To group subsets of pixels into bouton-like structures, we introduced a proximity matrix and optimized the clustering configuration by the Markov Chain Monte Carlo method with Simulated Annealing algorithm, which is established in statistical physics. We applied this procedure to the neuropil in the isolated central nervous system of Drosophila larvae and succeeded in extracting individual bouton-like structures and relating them to population neural activity for larval fictive motion. These results demonstrate the usefulness and significance of the neuropil calcium imaging and its decomposition to study the spatiotemporal patterns during the operation of neural processing.