Objective: Focal cortical dysplasias (FCDs) are a common cause of drug-resistant focal epilepsy and are often difficult to visually identify on MRI. Automated FCD detection is challenging, at least in part due to variability of FCD lesions as well as normal cortex. We implemented a normative modeling approach based on a novel set of local imaging features, allowing for estimates of similarity and automated detection of FCD lesions. Methods: Standardized MPRAGE, T2 and FLAIR MR images were obtained in 15 patients with radiologically or histologically diagnosed FCDs and 30 healthy volunteers. A set of generic 3D multiscale local image filters were computed for each MR contrast then sampled onto the gray-white junction surface. We created a latent representation of normal cortical variability in healthy volunteers, allowing for outlier detection and estimates of similarity. Following local normalization, we automatically detected FCD lesions as outliers based on projection onto the mean FCD vector. Results: In our normative model, normal atypical cortical regions such as primary sensorimotor and paralimbic regions are outliers, as are FCDs. The anterior insula and mesial temporal cortex have similar features to FCDs. Following local normalization, we automatically detected FCDs with a sensitivity of 80% and specificity of 70%. Significance: Our normative modeling approach allows for identification of known atypical regions of normal cortex as well as FCD lesions. FCDs appear similar to some normal paralimbic and sensorimotor cortices, becoming more easily distinguished following local normalization. Automated detection of FCD lesions can help with presurgical planning. Our method for estimating similarity is generic and can be extended to identification of other types of lesions or normal cortical areas.