Flexible pressure sensors have found widespread application in medical diagnosis and human–computer interaction. These practical applications often require large-area pressure sensing arrays (PSAs) to explore spatial pressure distributions. However, the development of high-performance PSAs still presents significant challenges. Here, a new method is proposed for the fabrication of a high-resolution capacitive PSA in a 10 × 10 configuration. The array comprises the upper and lower electrode arrays of laser-induced graphene patterned on polyimide films, as well as the sandwiched dielectric units based on electrospun nanofiber membranes of H3PO4@PVPP (cross-linked polyvinyl pyrrolidone). Experimental results indicate that the PSA exhibits high sensitivity across a wide pressure range (0–200 kPa), while offering a rapid response/recovery time of 24/41 ms (at 1 kPa), an exceptionally low detection limit of ∼1.2 Pa, and a good stability of 4000 cycles. Moreover, the crosstalk of the PSA was found to be as low as 1.3%. By employing the PSA in conjunction with a convolutional neural network algorithm, a human foot-arch diagnostic system was developed for accurate detection and quantification of the plantar pressure for different types of foot arches. Such technology has significant potential in clinical detection and disease diagnosis of human foot.