Hyperion images are currently processed to level 1a (from 0 or raw data). These files of radiometrically corrected data in units either watts/(sr /spl times/ micron m/sup 2/) 40 for VNIR bands 80 SWIR bands. Each distributed image tape contains a log file, called "(EO-1 identifier).fix.log", that reports the bad corrupted pixels (called known pixels) found during pre-flight checking, and details how they were fixed. All should be image. However, still evident. In addition, there dark...

With the rapid growth of medical image data, it has become a current research hotspot that how to realize large amounts real-time upload and storage images with limited network bandwidth space. However, currently, compression technology cannot perform joint optimization rate (the degree compression) distortion (reconstruction effect). Therefore, this study proposed algorithm based on variational autoencoder. This takes as common goal uses residual module directly transmit information, which...

At present, researchers have made great progress in the research of object detection, however, these studies mainly focus on detection images under normal lighting, ignoring target low light. And fields automatic driving at night and surveillance are usually obtained low-light environments. These problems such as poor brightness, contrast, obvious noise, which lead to a large amount information loss image. performance light is reduced. In this paper, we propose image enhancement method based...

Hyperion, a hyperspectral sensor, and the Advanced Land Imager (ALI) are carried on NASA's EO-1 satellite. The Evaluation Validation of for Sustainable Development (EVEOSD) is our project supporting mission. With 10% world's forests second largest country by area in world, Canada has natural requirement effective monitoring its forests. Eight test sites have been selected EVEOSD, with seven one US. Extensive fieldwork conducted at four these sites. A comparison made forest classification...

The Zynq-7000 consists of two elements AP SoC Processing System (PS) and a Programmable Logic (PL) unit. allows the us to implement video processing algorithm either as software program running on based PS or hardware accelerator inside PL. In this paper, we design face location system through software/hardware co-design Zynq-7000. firstly constructs skin model calculates similarity image data, then boundaries space projection finally utilizes Xilinx high-level synthesis tool Vivado HLS...

Aiming at the shortage of data transmission security and efficiency in wireless distributed transient pressure testing system, a new information source joint compression encryption scheme is proposed this paper by combining wavelet transform, semi-tensor product compressed sensing hash chain. Firstly, signals are decomposed into different components transform. Then, measurement matrix updated with chain related to plaintext, its used compress encrypt components. Finally, results previous...

Traditionally, the peak overpressures of multi-point shockwave are obtained through sensor array by using electrometric method, and then surface interpolation is carried out mathematical model to draw contour lines overpressure field.The cross-validation method mean absolute error (MAE), relative (MRE), root square (RMSE) was proposed in this study achieve high precision effective lines.The quantitative test Kinney-Graham Formula verified, a 7.62-mm caliber gun naval were based on polar...

In the process of transient test, due to insufficient bandwidth pressure sensor, test data is inaccurate. Firstly, based on projection shock tube signal in sparse domain, feature expression sample obtained. Secondly, problem solved by inverse modeling sensor dynamic compensation system swarm intelligence algorithm. this paper, method used compensate signals 85XX series sensors made Endevco company United States, working widened obviously, rise time can be compensated 12.5 μs, and overshoot...

Neural Network (NN) ensemble approach has been an appealing topic in the field of analog circuit fault diagnosis lately. In this paper, a new method for circuits with tolerance based on NN cross-validation is proposed. Firstly, bias-variance decomposition shows theoretical guide how to choose component networks when composing ensemble. Secondly, output voltage signal Circuit Under Test (CUT) obtained after stimulus imposed CUT. After getting corresponding feature sets, Bagging algorithm...

In order to solve the question of low recognition rate and complex identification process in software radio automatic modulation standard issue, technology Decision Tree was put forward. After studying characteristics digital signals feature parameters, then we need select appropriate tree structure, determine characteristic parameters which used on each non-end node, choose a suitable decision threshold. It has great influence results through analysing comparison with different order, get...

LXI(LAN Extensions for Instrument) distributed bus technology based on Ethernet had solved the problem of measurement at wide area. The test system consists LXI acquisition instrument which can meet shock wave signal test. In view characteristics field such as it has many points and long time sampling rate. We designed testing system. And paper describes detailedly composition hardware framework software. does frequency analysis to transient by wavelet transform get optimum decomposition...

In the process of digital micromirror device (DMD) mask projection lithography, lithography efficiency will be enhanced greatly by path planning pattern transfer. This paper proposes a new dual operator and population ant colony (DODPACO) algorithm. Firstly, load operators feedback are used to update local global pheromones in white colony, is yellow colony. The concept information entropy regulate number colonies adaptively. Secondly, take eight groups large-scale data TSPLIB as examples...

The shockwave signal is affected by the weapon launch and external environment, it often mixed with many kinds of noise, some even submerged. To detect extract under low signal-to-noise ratio, transient SNR, power-law detector higher-order c umulant spectrum (HOCS) Dual-tree complex wavelet transform (DTCWT) extraction model are proposed in study. average power noise different SNR was calculated comparing background instantaneous shockwave. Based on detection HOCS, two spectra analyzed....

This article introduced the process of image stitching based on SIFT and RANSAC. It was pointed out that reliability calculation RANSAC algorithm is an idealized assumption which does not hold. The false will lead to instability results, especially in multi-map or panorama situations. To solve problem, this paper presents a strategy limit range randomly points selecting. Experiments show can effectively reduce effect caused by random errors, obtain more stable result.

In the wireless distributed transient signal test field, redundant data is generated by conventional system in a long time and high sampling rate test. And it leads to fact that cannot be transmitted real-time. This paper uses characteristics of pre-process with Nyquist theorem real-time reduce data, ensures accurate same time, order achieve low-latency transmission. this method used process actual test, compression ratio 0.617% takes about 0.06%FS error. The results show algorithm can...

With the development of information technology, amount data signal acquisition end in high sampling rate fields such as wireless communication and medical imaging based on Nyquist theorem has increased dramatically. The resulting intensive processing brought great challenges to collection, storage, transmission system. Researchers began focus "sparse" characteristic signal, hoping see whole from small, develop a more efficient method. Recent studies showed that Sparse Fourier Transform...

In the compressed sensing (CS) reconstruction algorithms, problems of overestimation and large redundancy candidate atoms will affect accuracy probability algorithm when using Sparsity Adaptive Matching Pursuit (SAMP) algorithm. this paper, we propose an improved SAMP based on a double threshold, set reduction, adaptive backtracking methods. The uses threshold variable step-size method to improve sparsity judgment reduces undetermined atomic in small step stage enhance stability. At same...

A new non-coherent sequential code acquisition scheme is proposed based on the joint detection of correct cells (H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> cells). The comparisons ASN with that conventional cell-by-cell system show at almost same cost a significant improvement can be achieved and much more robust mean time performance against change phase offset variation or signal-to-noise ratio per chip (SNR/chip) reached....

Overcomplete dictionaries occupy an essential position in the sparse representation of signals. The dictionary construction method typically represented by <math xmlns="http://www.w3.org/1998/Math/MathML" id="M1"> <mi>K</mi> </math> singular value decomposition (KSVD) is widely used because its concise and efficient features. In this paper, based on background transient signal detection, adaptive estimation KSVD (ASE-KSVD) training proposed to solve redundant iteration problem caused fixed...