Belief change with noisy sensing in the situation calculus

Situation calculus has been applied widely in arti?cial intelligence to model and reason about actions and changes in dynamic systems. Since actions carried out by agents will cause constant changes of the agents’ beliefs, how to manage
these changes is a very important issue. Shapiro et al. [22] is one of the studies that considered this issue. However, in this framework, the problem of noisy sensing, which often presents in real-world applications, is not considered. As a
consequence, noisy sensing actions in this framework will lead to an agent facing inconsistent situation and subsequently the agent cannot proceed further. In this paper, we investigate how noisy sensing actions can be handled in iterated
belief change within the situation calculus formalism. We extend the framework proposed in [22] with the capability of managing noisy sensings. We demonstrate that an agent can still detect the actual situation when the ratio of noisy sensing actions vs. accurate sensing actions is limited. We prove that our framework subsumes the iterated belief change strategy in [22] when all sensing actions are accurate. Furthermore, we prove that our framework can adequately handle belief introspection, mistaken beliefs, belief revision and belief update even with noisy sensing, as done in [22] with accurate sensing actions only.

Privacy-Protected Facial Biometric Verification via Fuzzy Forest Learning

Although visual surveillance has emerged as an effective technolody for public security, privacy has become an issue of great concern in the transmission and distribution of surveillance videos. For example, personal facial images should not be browsed without permission. To cope with this issue, face image scrambling has emerged as a simple solution for privacyrelated applications. Consequently, online facial biometric verification needs to be carried out in the scrambled domain thus bringing a new challenge to face classification. In this paper, we investigate face verification issues in the scrambled domain and propose a novel scheme to handle this challenge. In our proposed method, to make feature extraction from scrambled face images robust, a biased random subspace sampling scheme is applied to construct fuzzy decision trees from randomly selected features, and fuzzy forest decision using fuzzy memberships is then obtained from combining all fuzzy tree decisions. In our experiment, we first estimated the optimal parameters for the construction of the random forest, and then applied the optimized model to the benchmark tests using three publically available face datasets. The experimental results validated that our proposed scheme can robustly cope with the challenging tests in the scrambled domain, and achieved an improved accuracy over all tests, making our method a promising candidate for the emerging privacy-related facial biometric applications.

Verification of Key Generation from Individual OFDM Subcarrier's Channel Response

This paper presents a key generation system derived from the channel response of individual subcarrier in orthogonal frequency-division multiplexing (OFDM) systems. Practical aspects of the security were investigated by implementing our key generation scheme on a wireless open-access research platform (WARP), which enables us to obtain channel estimation of individual OFDM subcarriers, a feature not currently available in most commercial wireless interface cards. Channel response of individual OFDM subcarrier is usually a wide sense stationary random process, which allows us to find the optimal probing period and maximize the key generation rate. The implementation requires cross layer design as it involves interaction between physical and MAC layer. We have experimentally verified the feasibility and principles of key generation, and also evaluated the performance of our system in terms of randomness, key generation rate and key disagreement rate, which proves that OFDM subcarrier's channel responses are valid for key generation.

Robust indoor speaker recognition in a network of audio and video sensors

Situational awareness is achieved naturally by the human senses of sight and hearing in combination. Automatic scene understanding aims at replicating this human ability using microphones and cameras in cooperation. In this paper, audio and video signals are fused and integrated at different levels of semantic abstractions. We detect and track a speaker who is relatively unconstrained, i.e., free to move indoors within an area larger than the comparable reported work, which is usually limited to round table meetings. The system is relatively simple: consisting of just 4 microphone pairs and a single camera. Results show that the overall multimodal tracker is more reliable than single modality systems, tolerating large occlusions and cross-talk. System evaluation is performed on both single and multi-modality tracking. The performance improvement given by the audio–video integration and fusion is quantified in terms of tracking precision and accuracy as well as speaker diarisation error rate and precision–recall (recognition). Improvements vs. the closest works are evaluated: 56% sound source localisation computational cost over an audio only system, 8% speaker diarisation error rate over an audio only speaker recognition unit and 36% on the precision–recall metric over an audio–video dominant speaker recognition method.

Time-resolved dosimetric verification of respiratory-gated radiotherapy exposures using a high-resolution 2D ionisation chamber array

The aim of this work was to track and verify the delivery of respiratory-gated irradiations, performed with three versions of TrueBeam linac, using a novel phantom arrangement that combined the OCTAVIUS® SRS 1000 array with a moving platform. The platform was programmed to generate sinusoidal motion of the array. This motion was tracked using the real-time position management (RPM) system and four amplitude gating options were employed to interrupt MV beam delivery when the platform was not located within set limits. Time-resolved spatial information extracted from analysis of x-ray fluences measured by the array was compared to the programmed motion of the platform and to the trace recorded by the RPM system during the delivery of the x-ray field. Temporal data recorded by the phantom and the RPM system were validated against trajectory log files, recorded by the linac during the irradiation, as well as oscilloscope waveforms recorded from the linac target signal. Gamma analysis was employed to compare time-integrated 2D x-ray dose fluences with theoretical fluences derived from the probability density function for each of the gating settings applied, where gamma criteria of 2%/2 mm, 1%/1 mm and 0.5%/0.5 mm were used to evaluate the limitations of the RPM system. Excellent agreement was observed in the analysis of spatial information extracted from the SRS 1000 array measurements. Comparisons of the average platform position with the expected position indicated absolute deviations of  <0.5 mm for all four gating settings. Differences were observed when comparing time-resolved beam-on data stored in the RPM files and trajectory logs to the true target signal waveforms. Trajectory log files underestimated the cycle time between consecutive beam-on windows by 10.0  ±  0.8 ms. All measured fluences achieved 100% pass-rates using gamma criteria of 2%/2 mm and 50% of the fluences achieved pass-rates  >90% when criteria of 0.5%/0.5 mm were used. Results using this novel phantom arrangement indicate that the RPM system is capable of accurately gating x-ray exposure during the delivery of a fixed-field treatment beam.