Differences in electrochemical characteristics among Li-ion batteries result in erroneous state-of-charge (SOC)/capacity estimation and state-of-health (SOH) prediction when using the existing dual extended Kalman filter (DEKF) algorithm. This paper presents a complementary cooperation algorithm based on DEKF combined with pattern recognition as an application Hamming neural network to the identification of suitable battery model parameters for improved SOC/capacity estimation and SOH prediction. Two kinds of pattern such as discharging/charging voltage pattern (DCVP) and capacity pattern (CP) were measured, together with the battery parameters, as representative patterns. Through statistical analysis, the Hamming network is applied for identification of the representative DCVP and CP that most closely matche that of the arbitrary battery to be measured. The model parameters of the representative battery are then applied for SOC/capacity estimation and SOH prediction of the arbitrary battery using the DEKF. This avoids the need for repeated parameter measurement.