On-line recognition differs from off-line recognition in that additional information about the drawing order of the strokes is available. This temporal information makes it easier to recognize handwritten texts with an on-line recognition system. In this paper we present a method for the recove y of the stroke order from static handwritten images. The algorithm was tested by classifying the words of an off-lane database with a state-of-the-art on-line recognition system. On this database with 150 different words, written by four cooperative writers, a recognition rate of 97.4% was obtained.

Moisture supply in the Pamir Mountains of Central Asia significantly determines the hydrological cycle and, as a result, impacts the local communities via hazards or socioeconomic aspects, such as hydropower, agriculture and infrastructure. Scarce and unreliable in situ data prevent an accurate assessment of moisture supply, as well as its temporal and spatial variability in this strongly-heterogeneous environment. On the other hand, a clear understanding of climatic and surface processes is required in order to assess water resources and natural hazards. We propose to evaluate the potential of remote sensing and regional climate model (RCM) data to overcome such issues. Difficulties arise for the direct analysis of precipitation if the events are sporadic and when the amounts are low. We hence apply a harmonic time series analysis (HANTS) algorithm to derive spatio-temporal precipitation distributions and to determine regional boundaries delimiting areas where winter or summer precipitation dominate moisture supply. We complement the study with remote sensing-based products, such as temperature, snow cover and liquid water equivalent thickness. We find a strong intra-and inter-annual variability of meteorological parameters that result in strongly variable water budget and water mobilization. Climatic variability and its effects on floods and droughts are discussed for three outstanding years. The in-house Remote Sens. 2015, 7 9728 developed HANTS toolbox is a promising instrument to unravel periodic signals in remote sensing time series, even in complex areas, such as the Pamir.

Off-line handwriting recognition has wider applications than on-line recognition, yet it seems to be a harder problem. While on-line recognition is based on pen trajectory data, off-line recognition has to rely on pixel data only. We present a comparison between an off-line and an on-line recognition system using the same databases and system design. Both systems use a sliding window technique which avoids any segmentation before recognition. The recognizer is a hybrid system containing a neural network and a hidden Markov model. New normalization and feature extraction techniques for the off-line recognition are presented, including a connectionist approach for non-linear core height estimation. Results for uppercase, cursive and mixed case word recognition are reported. Finally a system combining the on-and off-line recognition is presented.

No abstract

No abstract