The Event-Related Potential (ERP) P300 is an Electroencephalography (EEG) signal evoked by external stimuli, characterised by a positive deflection at the 300 ms after presentation of an interesting stimulus for the user. In literature, P300-based Brain-Computer Interface (BCI) has been implemented to translate the subjects intent for restoring communication and control functions. In order to detect the ERP of the EEG signals and taking into account low signal-to-noise ratio (SNR) of the P300 signal, multiple trial averaging has been widely used, which permit diminishing the random noise. However, the signal averaging technique requires processing multiple EEG trials, thus the ITR (Information Transfer Rate) of the P300-based BCI is significantly reduced. An open challenge of P300-based BCI systems focuses on recognizing ERP signals (as target and no-target) using a reduced number of trials with an enough classification rate. In this work, we propose a novel method based on Regularized Logistic Regression (RLR) using as features the coefficients obtained from a Canonical Correlation Analysis (CCA), as method for detection of visual P300 ERPs using a reduced number of EEG trials (1, 2, 3 and 4 trials). The proposed method was evaluated with a freely available EEG dataset and was compared with two widely used methods: Mean-Amplitude LDA and Step-Wise LDA. Finally, the proposed method outperforms standard algorithms for P300 detection, providing a maximum accuracy of 86.3% (p=0.27 to Mean Amplitude LDA and p less than 0.05 to Stepwise-LDA) and maximum Area Under Curve (AUC) of the Receiver Operating characteristic (ROC) of 0.753 (p=0.715 to Mean Amplitude LDA and p=0.228 to Stepwise LDA) on the recognition. Despite the performance of some metrics of proposed method are not significant, the results allow concluding that this one have an acceptable performance for the recognition of P300, using a reduced number of trials.