Tag: open-set recognition

A new open-set recognition network flags tomato leaf images from disease classes it was never trained on, outperforming standard baselines ...

A computer vision framework re-identifies visually uniform, unmarked pigs across an 18-day period with 88% mean daily accuracy, rising to ...

Researchers at Sun Yat-sen University have developed an improved convolutional prototype network that classifies known environmental sounds and rejects unknown ...

A low-cost EMG-controlled prosthetic arm uses confidence gating and temporal voting to refuse uncertain commands, cutting false movements by 75 ...

A new review of 29 recent studies finds that no deep learning intrusion detection system has been evaluated jointly for ...

A new federated learning framework called FedCAMP-IDS detects known and zero-day cyberattacks across heterogeneous IoT networks with up to 99.16 ...

A new vision–language framework called CLARiF combines diagnostic captioning, part-level segmentation, and retrieval-informed re-ranking to achieve state-of-the-art fine-grained pest recognition ...

A new study shows that adaptively fusing a closed-world classifier with an unsupervised open-set detector significantly improves detection of zero-day ...

A new survey of 161 studies maps how out-of-distribution data undermines AI-based mechanical fault diagnosis and argues that generalization, calibrated ...