Retina, a thin layer of tissue, is the essential part of the human visual system that receives light, converts it to neural signal, and transmits it to the brain for visual recognition. The red, green, and blue (R/G/B) cone retina cells are natural narrowband photodetectors (PDs) sensitive to R/G/B lights. Connecting with these R/G/B cone cells, a multilayer neuro-network in the retina provides neuromorphic pre-processing for R/G/B signals before transmitting them to the brain. Inspired by this sophistication, in this study we develop the narrowband (NB) PD imaging sensor that combines the R/G/B perovskite NB sensor array (mimicking the R/G/B cone cells) with a neuromorphic algorithm (mimicking the intermediate network of the retinal system) for high-fidelity panchromatic imaging. Compared to commercial image sensors, we utilize perovskite intrinsic NB PD to exempt the complex optical filter array and thus avoiding issues such as color aliasing, limited quantum efficiency and demosaicing processing. Furthermore, we utilize an asymmetric device configuration to spontaneously collect photocurrent under zero external bias, enabling a power-free photodetection feature. Results of the novel perovskite image sensor along with the machine learning algorithm assisted image information correction, demonstrate a promising device for efficient and intelligent panchromatic imaging.