Objective: To explore the optimal time for video-based electroencephalogram (EEG) monitoring and the detection rate of different seizure types in children with epilepsy. Methods: A total of 400 children with epilepsy admitted to First Hospital of Zhangjiakou City from January 2017 to January 2020 were selected. They were randomly divided into 8 groups, with 50 cases in each group. They were monitored by video-based EEG for 1, 2, 4, 6, 8, 10, 12 and 24 h. The EEG positive rate and EEG characteristics of epileptic children in different periods were recorded, and the EEG characteristics and positive detection of different seizure types were analyzed. Results: There were significant differences among groups at different monitoring time points (P<0.05). With the increase of monitoring time, the positive rate increased gradually, and the positive rate was 100% at 12 h. There was no significant difference between 6 and 12 h (P>0.05). Of the 360 children with positive detection, 64 had generalized tonic clonic seizures, 28 clonic seizures, 21 tonic seizures, 48 absence seizures, 29 myoclonic seizures, 34 atonic seizures, 43 simple partial seizures and 93 complex partial seizures. Patients with different attack types exhibited different clinical manifestations and EEG characteristics. The detection rate of generalized seizures was the highest at 6 hours of monitoring, and the detection rate of localized seizures was the highest at 12 hours of monitoring. Conclusion: Video-based EEG monitoring of children with epilepsy has good practical value. EEG characteristics can be used to judge the attack types of children. The positive rate is 100% after monitoring for 12 h, and the monitoring cost-effectiveness is the highest when monitoring for 6 h, which can basically meet the needs of clinical diagnosis.