Search

Diffusion models have demonstrated impressive performance in text-guided image generation. To leverage the knowledge of text-guided image generation models in image editing, current approaches either fine-tune the pretrained models using the input image (e.g., Imagic) or incorporate structure information as additional constraints into the pretrained models (e.g., ControlNet). However, fine-tuning large-scale diffusion models on a single image can lead to severe overfitting issues and lengthy inference time. The information leakage from pretrained models makes it challenging to preserve the text-irrelevant content of the input image while generating new features guided by language descriptions. On the other hand, methods that incorporate structural guidance (e.g., edge maps, semantic maps, keypoints) as additional constraints face limitations in preserving other attributes of the original image, such as colors or textures. A straightforward way to incorporate the original image is to directly use it as an additional control. However, since image editing methods are typically trained on the image reconstruction task, the incorporation can lead to the identical mapping issue, where the model learns to output an image identical to the input, resulting in limited editing capabilities. To address these challenges, we propose a text-to-image editing model with Image Information Removal module (IIR) to selectively erase color-related and texture-related information from the original image, allowing us to better preserve the text-irrelevant content and avoid the identical mapping issue. We evaluate our model on three benchmark datasets: CUB, Outdoor Scenes, and COCO. Our approach achieves the best editability-fidelity trade-off, and our edited images are approximately 35% more preferred by annotators than the prior-arts on COCO.

The purpose of this study was to assess the efficacy and safety of the combination of tirofiban with intravenous thrombolysis (IVT) in treating patients with capsular warning syndrome (CWS) who failed to respond to the treatment of intravenous thrombolysis alone. Tirofiban was approved for the treatment of CWS patients with fluctuating symptoms or no substantial improvement after intravenous thrombolysis within 24 h in our hospital from October 2019 to June 2021. Patients were evaluated with the National Institutes of Health Stroke Scale (NIHSS) at admission, at 72 h post-thrombolysis, at 1-week, and at 3-months with the modified Rankin Scales (MRS) score. A total of 12 patients received tirofiban and eight patients received control treatment with a history of CWS in our cohort. Among the patients, 13 patients smoked more than one pack of cigarettes a day, 17 had hypertension, 17 had hypercholesterolemia, 7 had diabetes, 1 had the history of cerebral infarction, 2 had atrial fibrillation, 7 had mild big vascular stenosis, 13 had lesions of the perforating branch by imaging, and 19 had acute capsular infarction. In both the tirofiban and control groups, NIHSS scores were significantly reduced after intravenous thrombolysis or 1-week after onset compared with before intravenous thrombolysis (P < 0.001). Before and after intravenous thrombolysis, there were no differences between the tirofiban group and control group (P = 0.970, P = 0.384, respectively). The tirofiban group, however, showed remarkably lower scores in both 1-week NIHSS and 3-month MRS than the control (P = 0.012, P = 0.003, respectively). Our study revealed that tirofiban did not increase the risk of hemorrhage and had favorable clinical efficacy as a remedial treatment for CWS patients with poor prognosis for intravenous thrombolysis, therefore indicating great potential for broader use.

Satellite laser ranging (SLR) had been operated at a pulse repetition frequency (PRF) from ∼10 Hz to 10 kHz; the ultra-high PRF of SLR (UH-SLR) is a trend of development. In this study, an alternate working mode of laser firing and gated pulse bursts is proposed to solve the problem of laser echo interference by laser backscattering. Through an ultra-high PRF of 200-kHz picosecond green laser with single-pulse energy of 80 μJ and a pulse width of 10 ps and a ranging gate device, UH-SLR has been built by an aperture of the 60-cm SLR system in the Shanghai Astronomical Observatory. By this UH-SLR, low-orbit to high-orbit and geostationary orbit satellites are measured night and day and also for low-orbit and medium-orbit satellites in the daytime. The normal point (NP) accuracy is ∼30 μm for low-orbit satellites and ∼100 μm for high-orbit satellites, which provides an effective method for the development of ultra-high PRF and high-precision space target laser ranging.

Because of the problems of low operation efficiency and poor energy management of multienergy input and output system with complex load demand and energy supply, this paper uses the double-layer nondominated sorting genetic algorithm to optimize the multienergy complementary microgrid system in real-time, allocating reasonably the output of each energy supply end and reducing the energy consumption of the system on the premise of meeting the demand of cooling, thermal and power load, so as to improve the economy of the whole system. According to the system load demand and operation mode, the first layer of this double-layer operation strategy calculates the power required by each node of the microgrid system to reduce the system loss. The second layer calculates the output of each equipment by using nondominated sorting genetic algorithm with various energy values calculated in the first layer as constraint conditions, considering the operation characteristics of various equipment and aiming at economy and environmental protection. In this paper, a typical model of energy input-output is established. This model combines with the operation control strategy suitable for multienergy complementary microgrid system, considers the operation mode and equipment characteristics of the system, and uses a double-layer nondominated sorting genetic algorithm to optimize the operation of each equipment in the multienergy complementary system in real time, so as to reduce the operation cost of the system.

Movie genre classification has been widely studied in recent years due to its various applications in video editing, summarization, and recommendation. Prior work has typically addressed this task by predicting genres based solely on the visual content. As a result, predictions from these methods often perform poorly for genres such as documentary or musical, since non-visual modalities like audio or language play an important role in correctly classifying these genres. In addition, the analysis of long videos at frame level is always associated with high computational cost and makes the prediction less efficient. To address these two issues, we propose a Multi-Modal approach leveraging shot information, MMShot, to classify video genres in an efficient and effective way. We evaluate our method on MovieNet and Condensed Movies for genre classification, achieving 17% ~ 21% improvement on mean Average Precision (mAP) over the state-of-the-art. Extensive experiments are conducted to demonstrate the ability of MMShot for long video analysis and uncover the correlations between genres and multiple movie elements. We also demonstrate our approach's ability to generalize by evaluating the scene boundary detection task, achieving 1.1% improvement on Average Precision (AP) over the state-of-the-art.

Image manipulation has attracted a lot of interest due to its wide range of applications. Prior work modifies images either from low-level manipulation, such as image inpainting or through manual edits via paintbrushes and scribbles, or from high-level manipulation, employing deep generative networks to output an image conditioned on high-level semantic input. In this study, we propose Semantic Image Manipulation with Background-guided Internal Learning (SIMBIL), which combines high-level and low-level manipulation. Specifically, users can edit an image at the semantic level by applying changes on a scene graph. Then our model manipulates the image at the pixel level according to the modified scene graph. There are two major advantages of our approach. First, high-level manipulation of scene graphs requires less manual effort from the user compared to manipulating raw image pixels. Second, our low-level internal learning approach is scalable to images of various sizes without reliance on external visual datasets for training. We outperform the state-of-the-art in a quantitative and qualitative evaluation on the CLEVR and Visual Genome datasets. Experiments show 8 points improvement on FID scores (CLEVR) and 27% improvement on user evaluation (Visual Genome), demonstrating the effectiveness of our approach.

Debris laser ranging (DLR) is receiving considerable attention as an accurate and effective method of determining and predicting the orbits of space debris. This paper reports some technologies of DLR, such as the high pulse repetition frequency (PRF) laser pulse, large-aperture telescope, telescope array, multi-static stations receiving signals. DLR with a picosecond laser at the Shanghai Astronomical Observatory is also presented. A few hundred laps of space debris laser-ranging measurements have been made. A double-pulse picosecond laser with an average power of 4.2 W, a PRF of 1 kHz, and a wavelength of 532 nm has been implemented successfully in DLR, it’s the first time that DLR technology has reached a ranging precision at the sub-decimeter level. In addition, the characteristics of the picosecond-pulse-width laser transmission with the advantages of transmission in laser ranging were analyzed. With a mode of the pulse-burst picosecond laser having high average power, the DLR system has tracked small debris with a radar cross-section (RCS) of 0.91 m2 at a ranging distance up to 1726.8 km, corresponding to an RCS of 0.1 m2 at a distance of 1000 km. These works are expected to provide new technologies to further improve the performance of DLR.

In recent years, the distributed energy supply system of the data center has become a hot topic in the industry, and the research on the coupling relationship between energy supply availability and economy is still insufficient. The thesis takes Hangzhou as an example to establish a mathematical model of coupling the energy supply cost and energy supply availability of the data center distributed system and checks it. The results show that the distributed energy system has the best economy and meets the energy supply availability requirements of the data center when the installation matching and system design are carried out according to the cooling and electric loads of the data center. In order to improve the reliability of the energy supply system in the data center, the prime mover of the gas distributed system should be equipped with one standby machine. At this time, the main limiting factor limiting the availability of the distributed energy supply system in the data center is the reliability of the gas supply network. The availability of two or more standby units does not significantly improve the availability of distributed energy supply system but increases the unit energy supply cost of the system.

International audience; Satellite laser ranging (SLR) stations in the International Laser Ranging Service (ILRS) global tracking network come in different shapes and sizes and were built by different institutions at different times using different technologies. In addition, those stations that have upgraded their systems and equipment are often operating a complementary mix of old and new. Such variety reduces the risk of systematic errors across all ILRS stations, and an operational advantage at one station can inform the direction and choices at another station. This paper describes the evolution of the ILRS network and the emergence of a new generation of SLR station, operating at kHz repetition rates, firing ultra-short laser pulses that are timestamped by epoch timers accurate to a few picoseconds. It discusses current trends, such as increased automation, higher repetition rate SLR and the challenges of eliminating systematic biases, and highlights possibilities in new technology. In addition to meeting the growing demand for laser tracking support from an increasing number of SLR targets, including a variety of Global Navigation Satellite Systems satellites, ILRS stations are striving to: meet the millimetre range accuracy science goals of the Global Geodetic Observing System; make laser range measurements to space debris objects in the absence of high optical cross-sectional retro-reflectors; further advances in deep space laser ranging and laser communications; and demonstrate accurate laser time transfer between continents.

Corner cube retroreflector (CCR) as a main optical component of laser ranging retroreflector array plays a key role in satellite laser ranging (SLR) to cooperative targets. To accurately estimate the echo energy of SLR, it is necessary to precompute photons’ distribution in the distance of SLR station by calculating the Far field diffraction pattern (FFDP) of CCR under various conditions. In this paper, the analysis of the effective reflection area and optical reflectivity of CCR for arbitrary incidence angle are carried out, in which a method to calculate the CCR effective reflection area with a wider applicability is used, and the difference between optical reflectivity of metal-coated CCR and uncoated (total internal reflection) CCR is also discussed. On this basis, combined with optical diffraction theory, a simulation program for CCR FFDP calculation is established, thereby producing FFDPs of CCR for a variety of incidence angles under different coating conditions. The results show that the FFDP of metal-coated CCR is almost unrelated to the azimuth angle of incident light or polarization, but is determined only by elevation angle of incident light. The pattern is always like Airy spot or its tensile deformation. Relatively, uncoated CCR’s FFDP has a more complex figure, its reflected energy is divided into several lobes whose size, number and position are all influenced by elevation and azimuth angle of incidence, and also by the polarization. Generally, the incidence direction which has a large total intensity of far optical field is to an extent the same as that of large effective reflection area and optical reflectivity. Furthermore, simulation results with uncoated CCR presents a higher relevance of incidence direction and FFDP.To verify the reliability and correctness of these simulation results, a diffraction optical experimental system at a wavelength of 1550 nm is set up to conduct laboratory confirmation, including laser, camera, beam expander, rotating platform and other essential optics. A silver-coated CCR and an uncoated CCR (both made of fuse silica, each has different dihedral angle offset) are chosen to measure their FFDPs with random polarization directions at several random incidence angles. All the measurement results are in good agreement with the simulations of FFDPs.

Avalanche photodiode single-photon detector is one of the ultra-sensitivity photoelectric detector, which has important applications in the fields of long-distance laser ranging, laser imaging, and quantum communication. However, due to the high temperature sensitivity of the avalanche voltage, the avalanche photodiode single-photon detector is prone to fluctuation of the avalanche gain when it works in the field environment, which leads to the delay drift and seriously reduces the time stability. In this paper, we proposed a method of stabilizing the delay of the single-photon detector. An embedded system was used to control avalanche photodiode at constant low temperature and compensate the delay drift of the detection circuit caused by the change of environment temperature in real time. A high time stability avalanche photodiode single-photon detector was realized by this method. In the experiment, the environment temperature changed from 16 ℃ to 36 ℃, and the avalanche photodiode was controlled at 15 ℃. After compensation, the delay drift of the avalanche photodiode single-photon detector was within ±1 ps, and the time deviation was 0.15 ps@100 s. This work is expected to provide an effective solution for the application of high-stability single-photon detector in the field and space environment.

Generating stylized captions for an image is an emerging topic in image captioning. Given an image as input, it requires the system to generate a caption that has a specific style (e.g., humorous, romantic, positive, and negative) while describing the image content semantically accurately. In this paper, we propose a novel stylized image captioning model that effectively takes both requirements into consideration. To this end, we first devise a new variant of LSTM, named style-factual LSTM, as the building block of our model. It uses two groups of matrices to capture the factual and stylized knowledge, respectively, and automatically learns the word-level weights of the two groups based on previous context. In addition, when we train the model to capture stylized elements, we propose an adaptive learning approach based on a reference factual model, it provides factual knowledge to the model as the model learns from stylized caption labels, and can adaptively compute how much information to supply at each time step. We evaluate our model on two stylized image captioning datasets, which contain humorous/romantic captions and positive/negative captions, respectively. Experiments shows that our proposed model outperforms the state-of-the-art approaches, without using extra ground truth supervision., 17 pages, 7 figures, ECCV 2018

International audience

Earthquakes can be detected by matching spatial patterns or phase properties from 1-D seismic waves. Current earthquake detection methods, such as waveform correlation and template matching, have difficulty detecting anomalous earthquakes that are not similar to other earthquakes. In recent years, machine-learning techniques for earthquake detection have been emerging as a new active research direction. In this paper, we develop a novel earthquake detection method based on dictionary learning. Our detection method first generates rich features via signal processing and statistical methods and further employs feature selection techniques to choose features that carry the most significant information. Based on these selected features, we build a dictionary for classifying earthquake events from non-earthquake events. To evaluate the performance of our dictionary-based detection methods, we test our method on a labquake dataset from Penn State University, which contains 3,357,566 time series data points with a 400 MHz sampling rate. 1,000 earthquake events are manually labeled in total, and the length of these earthquake events varies from 74 to 7151 data points. Through comparison to other detection methods, we show that our feature selection and dictionary learning incorporated earthquake detection method achieves an 80.1% prediction accuracy and outperforms the baseline methods in earthquake detection, including Template Matching (TM) and Support Vector Machine (SVM)., Comment: Workshop on Data Mining for Geophysics and Geology, SIAM International Conference on Data Mining 2018

Satellite laser ranging system calculates the distance from ground-based observatories to satellites using the round-trip travel time of laser pulse. The position of retro-reflectors on satellites needs to be corrected which is helpful to improvie the measuring precision of satellite laser ranging. The correction errors of center-of-mass(CoM)are mainly caused by the distribution effects of retro-reflectors on satellites.CoM is related to incident angle, structural alignment of retro-reflectors and ground-based position. Based on the reflecting probability of photons for retro-reflectors is proportional to the cross sections of retro-reflectors, the cross section area of corner reflectors is fitted and the probabilistic model is established using incident angle as the random variable. The corrections of CoMs of spherical satellite such as LAGEOS-1/2 are calculated and different CoM values are applied for SLR precise orbit determination using long-term full rate observation data with different WRMS results analyzed. At last, for the planar array retro-reflectors, the CoMs of BeiDou navigational satellite such as BeiDou-M3 are also calculated and analyzed using one month SLR full rate data. The result shows that the calculated CoMs based on probability theory have the comparative precision in SLR precise orbit determination.

To assess brain damage in Parkinson's disease (PD) based on apparent diffusion coefficient (ADC) data obtained from ultra-high b-values.Eighteen PD patients and 18 controls received diffusion-weighted imaging (DWI) with standard b-values (0, 1,000 s/mm(2)) and 15 b-values (0-5,000 s/mm(2)). Standard ADC (ADCst) maps were calculated from standard b-values, while maps of pure diffusion coefficients (D), pseudo-diffusion coefficients (D(*)), and ultra-high ADCs (ADCuh) were calculated from the 15 b-values using a tri-component model. In this model, D and D(*) values were quantified with a bi-exponential equation using b-values less than 2,000 s/mm(2), while ADCuh was quantified by fitting the signals at ultra-high b-values (2,000-5,000 s/mm(2)) to the mono-exponential equation. ADCst, ADCuh, D, and D(*) of the globus pallidus (GP), putamen (P), and substantia nigra (SN) were compared between PD patients and normal control subjects.ADCuh of the GP, P, and SN was significantly lower in PD patients than those in control subjects (P 0.001), while ADCst, D, and D(*) of the GP, P and SN were not different between the two groups (P 0.05).ADCuh may be a useful measurement for evaluating brain damage in PD patients.• DWI with ultra-high b-values may provide new insight into Parkinson's disease pathology • ADC calculated using ultra-high b-values is different between PD and controls • ADC uh may be associated with water transportation by aquaporins.

Parkinson's disease (PD) is a heterogeneous neurodegenerative disorder with variable clinicopathologic phenotypes and underlying neuropathologic mechanisms. Each clinical phenotype has a unique set of motor symptoms. Tremor is the most frequent initial motor symptom of PD and is the most difficult symptom to treat. The dentate nucleus (DN) is a deep iron-rich nucleus in the cerebellum and may be involved in PD tremor. In this study, we test the hypothesis that DN iron may be elevated in tremor-dominant PD patients using quantitative susceptibility mapping. Forty-three patients with PD [19 tremor dominant (TD)/24 akinetic rigidity (AR) dominant] and 48 healthy gender- and age-matched controls were recruited. Multi-echo gradient echo data were collected for each subject on a 3.0-T MR system. Inter-group susceptibility differences in the bilateral DN were investigated and correlations of clinical features with susceptibility were also examined. In contrast with the AR-dominant group, the TD group was found to have increased susceptibility in the bilateral DN when compared with healthy controls. In addition, susceptibility was positively correlated with tremor score in drug-naive PD patients. These findings indicate that iron load within the DN may make an important contribution to motor phenotypes in PD. Moreover, our results suggest that TD and AR-dominant phenotypes of PD can be differentiated on the basis of the susceptibility of the DN, at least at the group level. Copyright © 2016 John Wiley & Sons, Ltd.

In view of the severe steam channeling in the steam flooding of super heavy reservoir, lab experiment on steam channeling control were carried out. The combination of nitrogen foam and thermoset blocking agent was tested to seal steam channeling, in which thermoset blocking agent plugs big pore throats, while nitrogen foam adjusts steam absorption profile. The optimized foam formulation has a resistance factor of over 30 at 300 °C, can plug low oil saturation areas selectively, and applies to the plugging of high permeability zones in super-heavy oil reservoirs. Thermoset blocking agent, which would consolidate at 120 °C in 4 h and consolidate at 150 °C in 2 h, can provide effective plugging during dynamic steam flooding. The best steam channeling control mode was determined using parallel tube model. By the combination of nitrogen foam and thermoset blocking agent, the recovery rate is 5.7% higher than the application of nitrogen foam only, with the overall sweeping efficiency reaching up to 60.8%. In 2011, the mode was used in the steam flooding in Shan-56 reservoir. The water cut drops 10.2%, the wellhead temperature of producer drops more than 15 °C, the oil production of the well group increases over 28 tons per day, the valid period of a single cycle is up to 198 days, and the oil production increases 2 562 t, showing significant improvement in steam flooding. Key words: super heavy oil, steam flooding, nitrogen foam, thermoset blocking agent, steam channeling

This study compared the imaging quality, diagnostic accuracy, and apparent diffusion coefficient (ADC) values of reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) and full field-of-view (fFOV) single-shot echo-planar imaging with regard to patients with nonmuscle-invasive or muscle-invasive bladder cancer. Thirty-nine patients with 60 bladder tumors underwent rFOV and fFOV DWI in this internal review board-approved study. Pathologic and histologic grades were determined for all tumors. Two observers rated DWI image quality using a 4-point scale. Two radiologists who were blinded to the pathology findings reviewed 3 image sets (T2-weighted alone, T2-weighted plus fFOV DWI, and T2-weighted plus rFOV DWI) and assigned T stages and confidence levels for tumors of stage T2 or higher. The image quality scores for the 2 DWI sequences were assessed using the Wilcoxon signed-rank test. Differences in the diagnostic accuracy, sensitivity, and specificity for each image set were evaluated using the McNemar test. Differences in performance were analyzed by comparing the areas under the receiver-operating characteristic curves (ie, the Az values). A Mann–Whitney U test was used to compare the mean ADCs and the relationship between tumor stage and histologic grade. Image quality scores were significantly higher for rFOV (mean = 3.62) than for fFOV DWI (2.98; P

A 3D three-phase segmentally variable density perforation optimization model for horizontal wells is built by coupling reservoir fluid filtration, near wellbore inflow and wellbore conduit flow based on reservoir numerical simulation. The effects of 8 factors (filtration difference between heel/toe and middle intervals, imperforated interval, permeability heterogeneity, oil layer thickness heterogeneity, porosity heterogeneity, wellbore pressure drawdown, maximum perforation density, and perforation optimization principles) on perforation density and inflow profile, and that of fluid viscosity, casing diameter and pipe wall coarseness on well bore pressure drawdown, are analyzed for segmentally variable density perforation of horizontal wells. Results show that filtration difference between heel/toe and middle intervals, imperforated interval, permeability heterogeneity and oil layer thickness heterogeneity have significant effects on segmentally variable density perforation. In particular, different perforation density optimizations may occur when filtration difference between heel/toe and middle interval is not considered; imperforated interval may affect inflow profile; well bore pressure drawdown can be ruled out for segmentally variable density perforation of most horizontal wells onshore in China. The contrast between predicted and actual production of Well AT9-7H in the Tahe Oilfield indicates that the model is highly accurate. Key words: horizontal well, segmentally variable density perforation, reservoir numerical simulation, imperforated interval, inflow profile, heterogeneity

In Parkinson's disease (PD), iron elevation in specific brain regions as well as selective loss of dopaminergic neurons is a major pathologic feature. A reliable quantitative measure of iron deposition is a potential biomarker for PD and may contribute to the investigation of iron‐mediated PD. The primary purpose of this study is to assess iron variations in multiple deep grey matter nuclei in early PD with a novel MRI technique, quantitative susceptibility mapping (QSM). The inter‐group differences of susceptibility and R2(*) value in deep grey matter nuclei, namely head of caudate nucleus (CN), putamen (PUT), global pallidus (GP), substantia nigra (SN), and red nucleus (RN), and the correlations between regional iron deposition and the clinical features were explored in forty‐four early PD patients and 35 gender and age‐matched healthy controls. Susceptibility values were found to be elevated within bilateral SN and RN contralateral to the most affected limb in early PD compared with healthy controls (HCs). The finding of increased susceptibility in bilateral SN is consistent with work on a subgroup of patients at the earliest clinical detectable state (Hoehn and Yahr [1967]: Neurology 17:427–442; Stage I). However, increased R2(*) values were only seen within SN contralateral to the most affected limb in the PD group when compared with controls. Furthermore, bilateral SN magnetic susceptibility positively correlated with disease duration and UPDRS‐III scores in early PD. This finding supports the potential value of QSM as a non‐invasive quantitative biomarker of early PD. Hum Brain Mapp 36:4407–4420, 2015. © 2015 Wiley Periodicals, Inc.

The main purpose of this paper is to search formulas for different metals that relate the yield stress-strain with the strength coefficient and the strain-hardening exponent. For this purpose, the test data of nine alloys were used as basic data and the applicability of Hollomon’s equation at the yield point of the alloy was studied. This paper explores new equations relating the yield stress-strain with the strength coefficient and the strain-hardening exponent. At the same time, the study introduces a new fracture-ductility parameter. The new fracture-ductility parameter not only describes the applicability of the new equations to these alloys, but may also be better at describing the hardening behavior of a metallic material.

The Environmental Satellite (Envisat) mission was finished on April 8, 2012, and since that time, the attitude of the satellite has undergone significant changes. During the International Laser Ranging Service campaign, the Satellite Laser Ranging (SLR) stations have performed the range measurements to the satellite that allowed determination of the attitude and the spin period of Envisat during seven months of 2013. The spin axis of the satellite is stable within the radial coordinate system (RCS; fixed with the orbit) and is pointing in the direction opposite to the normal vector of the orbital plane in such a way that the spin axis makes an angle of 61.86 $^{\circ}$ with the nadir vector and 90.69 $^{\circ}$ with the along-track vector. The offset between the symmetry axis of the retroreflector panel and the spin axis of the satellite is 2.52 m and causes the meter-scale oscillations of the range measurements between the ground SLR system and the satellite during a pass. Envisat rotates in the counterclockwise (CCW) direction, with an inertial period of 134.74 s (September 25, 2013), and the spin period increases by 36.7 ms/day.

In this article, firstly the principles of r esource discovery mechanism and the techniques involved in Multi-Layer Overlay Network (MLON) based grid resource discovery mechanism are introduced; then, using Euclidean space distance and Top-K technique, the resource matching strategy and the transformation of queries in multidimensional attribute space are discussed in detail, and a resource discovery algorithm MLON-RSA (MLON-based Resource discovery Algorithm) is proposed and theoretically analyzed; at last, the feasibility, high efficiency, and high user satisfaction of the proposed algorithm are demonstrated through simulation tests.

In order to provide an objective and scientific theoretical basis for rational distribution of wheat growth in Yunnan Province, according to the relationship between Yunnan weather conditions and wheat growth adaptability, a study on eco-climate type regionalization of wheat growing areas in Yunnan was conducted using principal component analysis and GIS technology. The results show that Yunnan Province could be divided into four types, namely southern warm and humid wheat growing area, central semi-arid wheat growing area, central semi-humid wheat growing area and north-central cold wheat growing area.

In order to provide an objective and scientific theoretical basis for rational distribution of wheat growth in Yunnan Province, according to the relationship between Yunnan weather conditions and wheat growth adaptability, a study on eco-climate type regionalization of wheat growing areas in Yunnan was conducted using principal component analysis and GIS technology. The results show that Yunnan Province could be divided into four types, namely southern warm and humid wheat growing area, central semi-arid wheat growing area, central semi-humid wheat growing area and north-central cold wheat growing area.

A new data stream outlier detection algorithm SODRNN is proposed based on reverse nearest neighbors. We deal with the sliding window model, where outlier queries are performed in order to detect anomalies in the current window. The update of insertion or deletion only needs one scan of the current window, which improves efficiency. The capability of queries at arbitrary time on the whole current window is achieved by Query Manager procedure, which can capture the phenomenon of concept drift of data stream in time. Results of experiments conducted on both synthetic and real data sets show that SODRNN algorithm is both effective and efficient.