Robust classification approach for segmentation of blood defects in cod fillets based on deep convolutional neural networks and support vector machines and calculation of gripper vectors for robotic processing.

•Models based on deep CNN and SVM for classification of cod fillets are presented.•We present a robust pixel-wise classification model for automated blood segmentation.•Blood segmentation and localization in 3D space with gripper vectors is presented.•We visualize the effect of SVM hyperparameters on pixel-wise classification of blood.