Glaucoma is a chronic and progressive optic neurodegenerative disease leading to vision deterioration. The visual loss caused by glaucoma is irreversible and hence early detection of the disease is essential. A novel glaucoma detection system using digital fundus images is proposed based on hybrid features using a combination of fractal and textural features. Fractal features, namely fractal dimension (FD), lacunarity and correlation coefficient (CC), are used with second-order textural features for the effective detection of glaucoma. The feature set is then optimized by the sequential floating forward selection (SFFS) technique and the extracted features are fed as input to an adaptive neurofuzzy inference system (ANFIS) for classification of images as normal or abnormal. The proposed hybrid features achieved 96.8% specificity and 98% sensitivity with an accuracy of 97.45% and can be used in glaucoma mass screening.