Many occluded and ambiguous ground truths exist in object detection, making detectors unable to obtain optimal training samples. In this article, we revisit the suboptimal sample issue in label assignment for object detection and propose a novel detector-agnostic strategy, termed FSS, to address it. FSS reformulates label assignment as the process of selecting high-quality sub-optimal samples and progressively transforming them into optimal ones. Specifically, for each candidate, we estimate the probability of being an optimal sample by jointly considering localization quality and classification confidence, thereby constructing an instance-wise probability matrix. Based on the spatial distribution of potentially optimal samples, we introduce a Gaussian prior to adaptively determine the number of sub-optimal samples per instance. We then assign weights to these sub-optimal samples according to their optimality probabilities, enforcing consistent ranking between classification and localization and promoting the emergence of truly optimal samples. Extensive experiments on MS-COCO demonstrate the effectiveness and plug-and-play nature of FSS: when integrated into a modern one-stage detector, FSS achieves 50.8 AP under single-model, single-scale testing, without introducing any additional inference overhead.

The integration of cloud computing into traditional industrial control systems is accelerating the evolution of Industrial Cyber-Physical System (ICPS), enhancing intelligence and autonomy. However, this transition also expands the attack surface, introducing critical security vulnerabilities. To address these challenges, this article proposes a hybrid intrusion detection scheme for securing ICPSs that combines system state anomaly and network traffic anomaly detection. Specifically, an improved variation-Bayesian-based noise covariance-adaptive nonlinear Kalman filtering (IVB-NCA-NLKF) method is developed to model nonlinear system dynamics, enabling optimal state estimation in multi-sensor ICPS environments. Intrusions within the physical sensing system are identified by analyzing residual discrepancies between predicted and observed system states. Simultaneously, an adaptive network traffic anomaly detection mechanism is introduced, leveraging learned traffic patterns to detect node- and network-level anomalies through pattern matching. Extensive experiments on a simulated network control system demonstrate that the proposed framework achieves higher detection accuracy (92.14%) with a reduced false alarm rate (0.81%). Moreover, it not only detects known attacks and vulnerabilities but also uncovers stealthy attacks that induce system state deviations, providing a robust and comprehensive security solution for the safety protection of ICPS.