Search

In medicine, whether the fundus image is clear or not would affect the doctor's diagnosis of the disease. In this paper, the feature vector of clear or blurred image is constructed by the fundus image edge degree and the transformed energy of the wavelet coefficients. The fisher criterion is adopted in order to determine the optimal projection direction and boundaries of the classifier threshold to classify the clear image and fuzzy image. The method has certain reference for image clarity with similar image feature judgment.

Median filter is a popular image preprocessing way; it can inevitably introduce various noises and make image quality become bad in the process of image generation, collection and transmission. So we have to make a preprocessing procedure to restraint the image noise for the following process. Because median filter is a kind of nonlinear filtering, in practice, it may overcome the image details blurring comparing with a linear filter and can effectively filter the pulse interference and image scanning noise. But, as the large computational complexity, the image preprocessing is slow with software method. Alternatively, we can use hardware method to help the preprocessing. The field programmable gate array (FPGA) is the programmable component based on the list structure. It has abundant registers resources and can change logic function through the reconfiguration, which makes the flexibility of design increase greatly. In this paper, we adopt FPGA platform to achieve mid-value filtering algorithm through VHDL hardware describe language. In the condition of possibly using hardware resources, through improving the algorithm and optimizing the structure, it efficiently digs up the inner parallelism of algorithm and increases processing speed of filtering module. We adopt a quick median filtering algorithm and the algorithm can greatly save the hardware resources, so the data processing speed is quick.

The secondary instrument calibrator is mainly used for testing and calibrating the temperature or flux measuring meters used in engineering projects. In this paper, we mainly describe the design and implementation process of it. The Mcs51 microprocessor of the Intel Corporation, 5 – 1/2 bits A/D converter with micro-processor, and LCD module are used in our design. Furthermore, standard current source circuit and control circuit must be considered. On one hand, the calibrator can be used as a voltmeter or ammeter. On the other hand, it can provide high precision current source for the calibration of secondary instruments.

In this paper, a new design scheme of the grain measure and control system is put forward. The system is composed by the measurement control unit and data processing center. It can implement the real-time collection, communication and processing of the granary temperature and humidity. Temperature measurement accuracy can reach to ± 0.5 00°C, as humidity measurement accuracy ± 3% RH. The digital transmission of single bus is directly used for temperature and humidity, which can greatly reduce the cable number of the system, and improve the stability and anti-jamming of the system.

Image noise variance estimation is an important indicator to determine the image quality. In this paper, LMLSD algorithm based on Gaussian wave extraction is used for fundus image noise estimation, and it analyses the sub-block size effects for the performance of this algorithms. In the expriment, 6 different sub-block size are taken for the simulation and analysis of noise variance estimation respectively in 10 fundus images. Exprimental results show the accuracy of noise variance estimation which takes 5×5 sub-block size is better than others in 256*256 fundus images.

According to the actual demand of the system, the paper proposed scheme and implement for the stored-grain grossness remote monitoring system based on GPRS and Zigbee. The test result is shown that the max error is ∼0.02mA between inversion value of the server-side data and the measure value. The design should be served as the useful reference to the study of acquisition technology in similar system.

The paper proposed scheme and implement for the grossness inspection system of grain depot based on the actual demand of the system and the electrical confine analysis on the application environment. The test results is shown that, on the communication condition of 1Km distance and 5Kbps rate, the mistake code didn’t appeared and the measure error is less than 2.4% in the relative MSD, which can meet with the precision demand of the inspection system. The design should be served as the useful reference to the study of acquisition technology in similar system.

To need the requirement of measuring the grain quantity in the barn, the pressure sensors are mounted on the ground and the wall of the barn, and the grain quantity can be calculated by using these measurement values. To realize the data transmitting reliability and real time, the design proposal of the measurement node of the CAN-bus by using the chips of AT89C51 and the SJA1000 is put forward, and the design of software and the hardware of the node is given. To verify the effectivity and the reliability of the measurement node, some experiments are done, the results shown that, the measurement error is no more than 3% under the conspicuous level is equal to 0.05, the grain quantity can be measure by this system completely.

Based on the properties of transient feeble optical signal, which is the wide dynamic range and the feeble lower limit signal, it is presented a novel detecting scheme compressing signal real-time in logarithmic domain in this paper. After the principle analysis of the scheme, it has been implemented to choose appreciate parts, design the circuitry and test. The results indicated that the approach can make transient optical signal ranged 120dB compress in the range of 0~5V. And together with the circuitry design simplified and the dissipation decreased, the troubles at the back data acquisition module can be lessened greatly. It may be a novel clue for research on the tech of detecting the transient light and provide certain reference values to similar system design.

In order to analyze the influence of the satellite jitter on the image quality of the remote sensing camera, the image motion introduced by the low frequency sine jitter was given. The influence of the image motion along the roll, pitch and yaw direction on the image quality was studied chiefly. Finally, the degradation of the image quality caused by the jitter was computed concretely according to the datum come from two kinds of the emulation on the satellite kinetic jitter. The results showed that, with the increase of TDICCD integration number, the requirement of remote sensing camera for jitter amplitude is more stringent, and must be limited.

In this study, we present an integration scheme of preprocessing circuit for detecting optical pulse based FPGA-tech. Using the photoelectric detection technique, the scheme includes a the photodiode detector (PDD), which can convert an optical pulse to a small current signal; and a FPGA produces code to control MUXes for different channel, which changes circuit parameters to adapt different detection situation. Design issues are addressed. Testing circuits are implemented. Simulation wave displayed on the oscillograph and the data of the system acquiring are presented. And the feasibility of the scheme is confirmed by the data analyse. Presently, the scheme has been used in a lightning detection system.

Extracting the clear fundus images information is a prerequisite for diagnosing the patients’ conditions exactly in clinic. This paper adopts respectively de-noising algorithm based on multi-scale and that based on median filter to process real fundus images. It gives the LMLSD algorithm based on Gaussian wave extraction, which is used to estimate SNR (Signal-to-Noise) of the processed images and give an objective quantitative evaluation to treatment effect of the above two de-noising algorithms. The results show that the image processed by multi-scale algorithm is much clearer, and SNR increases significantly. The results of objective quantitative evaluation are in accordance with subjective visual feeling, so the selecting of evaluation algorithm is appropriate.

—A method for threshold selection to detect edge width of objects in the image was presented. Three order spline wavelet was choosed to complete multiscale wavelet edge detection and wavelet edge module values were obtained. After properties of these data was analyzed, a method for threshold selection was proposed using half of maximum of wavelet edge module values. Edge width of objects in the image was given a definition and can be calculated by area and girth of the transition edge using the threshold selected by the method proposed. Experiment results show that the algorithm is effective and performs better.