Search results

The mainly current processing tasks of satellite are the functions of the image pre-processing mass data storage and format conversion, data compression and forwarding, automatic analysis of the raw data and the target feature extraction and etc, which are trying to rely on independent processing of satellite, to minimize ground intervention.
ORASIS can carry out the satellite hyper-spectral data for real-time automatic data analysis, feature extraction and 10:1 to 20:1 data compression and real-time tactical downlink observations and results from the satellite delivered directly to the battlefield.
IOBP directly from the payload controller sustained receive data with the speed of up to 160Mbit / s.
Conclusions With the development of sensor, remote platform, data communication technique, modern remote sensing technology has entered a phase which can provide global observation data dynamically, rapidly, accurately and multi-means.
So far, some of developed countries successfully manufactured space-borne remote data real-time processing system, realized functions such as pre-processing on satellite, massive data storage and format conversion, data compression and forwarding, automatically analyze raw data and target feature extraction and so on, widely applied in meteorology, resource, disaster mitigation and military spy area.

Analysis of Dynamic Characteristics Dynamic responses of bridge-subgrade transition section in existing railway line are tested in different axial load and running speed of freight train and passenger train respectively, which can be used to analyse and master the adaptability of typical transition structure and accumulate data for further study of the dynamic characteristics of the bridge-subgrade transition section.
The test data of track structure stiffness are shown in Fig. 4.
The highest reduction ratio of the wheel load is 0.78 in all of the experimental transition sections, which is beyond the scope of the criterion.
But with a combination of situations on speed-raising in existing railway line in Japan and Germany and actual practices of limit value of derailment test in China, it still meets the requirement that the dynamic reduction ratio of wheel load is less than or equal to 0.8.
The major conclusions can be summarized as: According to the data collected by experimental test or track inspection car, the state of the railway line can be predicted and assessed by the simulation program.

Surface waviness were estimated by Fast Fourier Transfer from surface data measured using the Optiflat interferometer.
A one to two orders reduction in surface roughness was observed after first 8 minutes.
Consequently, polishing with a higher pressure might be helpful for the reduction of surface roughness when a softer polishing pad was used.
In this study, surface waviness was estimated by filtering the data of surface form with spatial wavelength range of 0.4 mm ~5.0 mm and making an average of those data within the wavelength range.
A two time reduction in surface waviness was observed in the first 8 minutes.

Load distribution, i. e. the distribution of reduction of each stand, is the core content of setting rolling regulations.
(4) Where, 0.21 0.3m h = + is a parameter obtained from a large number of experimental data.
On the basis of load distribution using Φ, the shape stiffness of the rolling mill and the rolled piece, the variation of mechanical crown, the variation of rolling force and the adjustment of reduction are calculated by the deviation of strip crown measured at the mill exit.
The values of adjusting parameters and on-line measured data are presented in Table 2.
Table 2 On-line measured data when objective adjustment of exit crown is -0.020mm （steel grade Q235, entry thickness 31.230mm, target thickness 3.500mm） stand No. 1 2 3 4 5 6 real exit thickness after adjusting rolling regulation h1[mm] 18.266 10.585 6.691 4.956 4.074 3.536 real rolling force P[kN] 18842 17774 16378 12036 9420 7207 variation of rolling force P∆ [kN] -834 -723 -656 variation of crown [mm] -0.019 Comparison of the RRRResults.

A sampling point location is noted with GPS.We acquire many chemical element’s concentration data in every sampling point with special instrument testing and analysing.
We mark every metal element’s concentration data.We calculate with interpolation function(Griddata).
According to eight heavy metal elements’ concentration in every sampling point,we obtain eight heavy metal elements’ average content of every area （）by excel’s data statistics.
K is assumed to be contamination rate of reduction.
Ingoing volumetric contamination quantity of unit time is: Outgoing volumetric contamination quantity of unit time is: If volumetric contamination occur primary reduction reaction,That reduction lead to change contamination quantity is .

Because EDM is heat energy machining, the reduction of the magnetic flux density must be considered.
On the other hand, when the nonmagnetized material was machined using EDM, the inversion of the polarity did not occur at any machining depth, and the reduction of the magnetic flux density depended on the hole depth.
Therefore, the measured values were lower than the surface data shown in Fig. 1.
The data from the Tesla meter was logged at a sampling interval of 0.5 s during machining.
However, clearly identifying the source of the variation based on the limited data is not possible.

The lines drawn refer to the best fit of the experimental data to a numerical solution of the continuity equation with sf as unknown parameter.
The symbols refer to the data measured, the lines drawn to fits with the simple model explained in the main text.
As the treatments used act only on the surfaces (with a possible exception for the deposition of Si3N4) the change of the decay behaviour is due to changes of s: The largest reduction of s is observed in a I2/ethanol solution, a somewhat smaller reduction after Si3N4 deposition at both surfaces and the smallest reduction after a HF dip.
This yields: HF dip 2.5*103cms-1 < s < 3.5*103cms-1 Si3N4 covered 7*10 1cms-1 < s < 3.5*102 cms-1 In I2/ethanol solution 0 < s < 2.5*10 2cms-1 It is clear that a HF dip does not suffice to reduce s and increase τs enough to neglect (τs)-1 in Eq.3 so that the measured decay time τeff corresponds to τb: If this is assumed the data in Fig.3 give τb=6.5µs, although the measurement in I2/ethanol solution shows that τb is at least 50µs.

The result shows that the total center distance obtained is 2609mm,which is decreased by 400mm compared with the original data, the other parameters are shown in Table 1.
To use the micro geometry study in ROMAX , using Monte Carlo method,the maximum transmission error under the load spectrum as the optimization target, the tooth lead crown modification as parameters variables, the tooth lead slope as the constraint condition, then the optimized modification scheme is obtained.The final optimum amount of modification is 355.46um and lead slope is 47.8 um. 4.3 Data analysis In order to reduce the deformation of the shafts, increasing the diameter of the shaft where gears installation, adding the shoulder fillets to avoid stress concentration, and the optimized results are shown in Fig.4 and Fig.5.
Both the simulated and experimental results show that the presented precise TPM(tooth profile modification) of helical gears is effective on vibration reduction around the working load, and the dynamic contact simulation is effective in estimating the effect of the TPM on vibration reduction in the designing stage.
A large reduction in the transmission error can effectively reduce the vibration and noise of the whole transmission system.
(1) Using mathematical software, in order to satisfy the requirements of the strength of the allowable value, the excavator has to decrease its total center distance to 2609mm, reduced by 400mm compare with the original data, improving the space usage

The low frequency sub-band images abstracted by DWT not only retain the main characteristics of the original image, but have the less data dimension compared with the original image[[] M.Zhu and A.M.Martinez:discriminant analysis.IEEE Trans.Pattern Analysis and Machine Intelligence,Vol.28(8),p1274-p1286,2006. ].
After passing a series of transitions, the image dimension number is reduced, largely reducing the disposing data amount of recognition dispose machine, and decreasing the noise and redundant message[[] Yuan Weilin,Ma yan,Liu Shengwei: " Analysis of Remote Sensing image compression based on DWT and Matlab".Geomatics World 2013(3) ]
Usually speaking, in order to satisfy the accuracy requirement, k shall be no less than 5, that is to say, every image will require at least 320 pieces of data to display.
Due to the large data size, the matching rate during the recognition process is slow, and the running time is too long.
However, the paper has not given detailed description in terms of the training classifier, and excessive data is filtered during the DWT process.

This is a nonlinear, non-parameter recognition algorithm, by defining a operation on the basis of the probability density function for each data on the level of the training set data and optimized kernel width parameters.
And for a typical, the number of neurons is equal to the number of training data model.
One of the important characteristics of MRI data sets is that neighboring voxels possess similar feature values, and the probability that they belong to the same cluster is great.
Second, since the trained SOM map serves as a good representative of the training samples, its use makes the PNN classifier more robust in the presence of noise data in the original training set.
We corrupt the data with synthetic Rician noise.