Abstract:Handwritten signatures are widely utilized as a form of personal recognition. However, they have the unfortunate shortcoming of being easily abused by those who would fake the identification or intent of an individual which might be very harmful. Therefore, the need for an automatic signature recognition system is crucial. In this paper, a signature recognition approach based on a probabilistic neural network (PNN) and wavelet transform average framing entropy (AFE) is proposed. The system was tested with a wavelet packet (WP) entropy denoted as a WP entropy neural network system (WPENN) and with a discrete wavelet transform (DWT) entropy denoted as a DWT entropy neural network system (DWENN). Our investigation was conducted over several wavelet families and different entropy types. Identification tasks, as well as verification tasks, were investigated for a comprehensive signature system study. Several other methods used in the literature were considered for comparison. Two databases were used for algorithm testing. The best recognition rate result was achieved by WPENN whereby the threshold entropy reached 92%.

No abstract

In many practical scenarios, targets tend to have certain mobility trends such as following a traverseable terrain, having a common starting/destination locations, or moving in a region with abundant resources. This work is interested in exploring the possible gain from sensor relocation in improving the localisation accuracy of targets that follow mobility trends similar to those previously observed. This objective is tackled using a three‐phase approach. In the first phase, the wireless sensor network tracks the targets based on the initial deployment. The second phase uses the location estimates from phase 1 to form a region of interest (ROI). The last phase carries out the sensor relocation to the ROI. Two fitness functions are explored for optimising sensors’ locations in the ROI, namely geometric dilution of precision and K‐coverage. K‐coverage offered the best performance especially for sensors with a short‐to‐medium detection range. The uniform random relocation offered a comparable performance with a relatively low computational complexity. Results also revealed the degradation in coverage rate due to relocating sensors to the ROI, and how optimising sensor locations outside the ROI can help in mending coverage holes.

No abstract