Ship detection and extraction using visual saliency and histogram of oriented gradient

A novel unsupervised ship detection and extraction method is proposed. A combination model based on visual saliency is constructed for searching the ship target regions and suppressing the false alarms. The salient target regions are extracted and marked through segmentation. Radon transform is applied to confirm the suspected ship targets with symmetry profiles. Then, a new descriptor, improved histogram of oriented gradient(HOG), is introduced to discriminate the real ships. The experimental results on real optical remote sensing images demonstrate that plenty of ships can be extracted and located successfully, and the number of ships can be accurately acquired. Furthermore, the proposed method is superior to the contrastive methods in terms of both accuracy rate and false alarm rate.

An Ensemble Approach for Emotion Cause Detection with Event Extraction and Multi-Kernel SVMs

In this paper, we present a new challenging task for emotion analysis, namely emotion cause extraction.In this task, we focus on the detection of emotion cause a.k.a the reason or the stimulant of an emotion, rather than the regular emotion classification or emotion component extraction. Since there is no open dataset for this task available, we first designed and annotated an emotion cause dataset which follows the scheme of W3 C Emotion Markup Language. We then present an emotion cause detection method by using event extraction framework,where a tree structure-based representation method is used to represent the events. Since the distribution of events is imbalanced in the training data, we propose an under-sampling-based bagging algorithm to solve this problem. Even with a limited training set, the proposed approach may still extract sufficient features for analysis by a bagging of multi-kernel based SVMs method. Evaluations show that our approach achieves an F-measure 7.04%higher than the state-of-the-art methods.

Rapid, simple and selective determination of 2,4-dinitrophenol by molecularly imprinted spin column extraction coupled with fluorescence detection

A rapid, simple and selective method based on molecularly imprinted, spin column extraction coupled with fluorescence detection was successfully established for the determination of 2,4-dinitrophenol in serum samples. The 2,4-dinitrophenol imprinted polymers exhibited highly selective recognition for the template molecule and the maximum adsorption capacity was 138.9 mg/g. The results indicated that when water is used as the loading solution, only 2,4-dinitrophenol could be adsorbed on the spin column without the remaining structural analogs（2-nitrophenol, 4-nitrophenol and phenol）. After eluting with acetonitrile/acetic acid（9/1, v/v）, 2,4-dinitrophenol in serum samples could be determined by using the fluorescence spectrometer, based on the fluorescence enhancement of fluorescein by the template molecule. Under the optimal conditions, the spiked recovery ranged from 95.8% to 103.4% and the detection limit was 1 nmol/L. The results confirmed the reliability and practicality of the protocol and revealed a good perspective of this method for biological sample analysis.

RS analysis of glaciers change in the Heihe River Basin, Northwest China, during the recent decades

The Heihe River Basin is the second largest inland river basin in Northwest China and it is also a hotspot in arid hydrology, water resources and other aspects of researches in cold regions. In addition, the Heihe River Basin has complete landscape, moderate watershed size, and typical social ecological environmental problems. So far, there has been no detailed assessment of glaciers change information of the whole river basin. 1：50,000 topographic map data, Landsat TM/ETM＋ remote sensing images and digital elevation model data were used in this research. Through integrated computer automatic interpretation and visual interpretation methods, the object-oriented image feature extraction method was applied to extract glacier outline information. Glaciers change data were derived from analysis, and the glacier variation and its response to climate change in the period 1956/1963–2007/ 2011 were also analyzed. The results show that：（1） In the period 1956/1963–2007/2011, the Heihe River Basin＇s glaciers had an evident retreat trend, the total area of glaciers decreased from 361.69 km2 to 231.17 km^2; shrinking at a rate of 36.08%, with average single glacier area decrease 0.14 km^2; the total number of the glaciers decreased from 967 to 800.（2） Glaciers in this basin are mainly distributed at elevations of 4300–4400 m, 4400–4500 m and 4500–4600 m; and there are significant regional differences in glaciers distribution and glaciers change.（3） Compared with other western mountain glaciers, glaciers retreat in the Heihe River Basin has a higher rate.（4） Analysis of the six meteorological stations＇ annual average temperature and precipitation data from 1960 to 2010 suggests that the mean annual temperature increased significantly and the annual precipitation also showed an increasing trend. It is concluded that glacier shrinkage is closely related with temperature rising, besides, glacier melting caused by rising temperatures greater than glacier mass supply by increased precipitation to some extent.

Extracting signal via stochastic resonance in the semiconductor optical amplifier

The stochastic resonance based on optical bistability in the semiconductor optical amplifier is numerically investigated to extract a weak pulse signal buried in noise. The output property of optical bistability under different system parameters is analyzed, which determines the performance of the stochastic resonance. Through optimizing these parameters, the noise-hidden signal is extracted via stochastic resonance, in which the maximum cross-correlation gain higher than nine is obtained. This provides a novel technology for detecting a weak optical signal in various signal processing fields.