Search

In order to reduce the pollution of the environment, China has implemented a new ultra-low emission control regulations for polluting gas, requiring new coal-fired power plant emissions SO2 less than 30ppm, NO less than 75ppm, NO2 less than 50ppm, Monitoring low concentration of NO and SO2 mixed gases , DOAS technology facing new challenges, SO2 absorb significantly weaken at the original absorption peak, what more the SNR is very low, it is difficult to extract the characteristic signal, and thus cannot obtain its concentration. So it cannot separate the signal of NO from the mixed gas at the wavelength of 200~230nm through the law of spectral superposition, it cannot calculate the concentration of NO. The classical DOAS technology cannot meet the needs of monitoring. In this paper, we found another absorption spectrum segment of SO2, the SNR is 10 times higher than before, Will not be affected by NO, can calculate the concentration of SO2 accurately, A new method of segmentation and demagnetization separation technology of spectral signals is proposed, which achieves the monitoring the low concentration mixed gas accurately. This function cannot be achieved by the classical DOAS. Detection limit of this method is 0.1ppm per meter which is higher than before, The relative error below 5p when the concentration between 0∼5ppm, the concentration of NO between 6∼75ppm and SO2 between 6∼30ppm the relative error below 1.5p, it has made a great breakthrough In the low concentration of NO and SO2 monitoring. It has great scientific significance and reference value for the development of coal-fired power plant emission control, atmospheric environmental monitoring and high-precision on-line instrumentation.

In this paper, a new BCE with multi-focusing microlens array (MLA) was proposed. The BCE consist of detachable micro-hole array (MHA), multi-focusing MLA and spherical substrate, thus allowing it to have a large FOV without crosstalk and stray light. The MHA was fabricated by the precision machining and the parameters of the microlens varied depend on the aperture of micro-hole, through which the implementation of the multi-focusing MLA was realized under the negative pressure. Without the pattern transfer and substrate reshaping, the whole fabrication method was capable of accomplishing within several minutes by using microinjection technology. Furthermore, the method is cost-effective and easy for operation, thus providing a feasible method for the mass production of the BCE. The corresponding image processing was used to realize the image stitching for the sub-image of each single microlens, which offering an integral image in large FOV. The image stitching was implemented through the overlap between the adjacent sub-images and the feature points between the adjacent sub-images were captured by the Harris point detection. By using the adaptive non-maximal suppression, numerous potential mismatching points were eliminated and the algorithm efficiency was proved effectively. Following this, the random sample consensus (RANSAC) was used for feature points matching, by which the relation of projection transformation of the image is obtained. The implementation of the accurate image matching was then realized after the smooth transition by weighted average method. Experimental results indicate that the image-stitching algorithm can be applied for the curved BCE in large field.

Elderly care becomes more and more prominent in China as the population is aging fast and the number of aging population is large. Falls, as one of the biggest challenges in elderly guardianship system, have a serious impact on both physical health and mental health of the aged. Based on feature descriptors, such as aspect ratio of human silhouette, velocity of mass center, moving distance of head and angle of the ultimate posture, a novel vision-based fall detection method was proposed in this paper. A fast median method of background modeling with three frames was also suggested. Compared with the conventional bounding box and ellipse method, the novel fall detection technique is not only applicable for recognizing the fall behaviors end of lying down but also suitable for detecting the fall behaviors end of kneeling down and sitting down. In addition, numerous experiment results showed that the method had a good performance in recognition accuracy on the premise of not adding the cost of time.

In industrial production, machine vision is widely used in a variety of automated testing industry, because of its non-contact, high precision, fast detection time, etc. This paper introduces a nondestructive detection system for the outline dimensions of a kind of special workpiece. For the purpose of controlling the quality of image，an ultra-high-resolution monochrome CCD image sensor is adopted to capture source image. Then it’s image preprocessing, including image clipping processing, grey scale processing, image denoising, etc. Then it was the progress of edge extraction, using gradient operator to get its contour edge, then a piecewise fitting method to get the dimensions. This method achieves once measuring multiple outline dimensions without moving the workpiece. It has been used in industrial production. The result sees its practical value for meeting the production needs, with the advantages of high accuracy, fast speed high stability of the measurement system through experiments.

A novel multi-focusing curved microlens array (MLA), inspired by compound eyes, was fabricated using a homebuilt mold in this paper. By gas-assist deformation, and using the PDMS film as the template, MLA with variable focal length were fabricated through the different apertures of the micro-stomatas attached on the mold. Fabricated in one step, without reverse mould process, makes the whole fabrication process simple, cost-effective and deformation less. The ultraviolet curing adhesive NOA81, which was cured in seconds, has excellent optical performance and a wide operating temperature range. The optical performances of the MLA were demonstrated by optical experiments and simulation.

In the near infrared image containing complex background, eye detection is a great challenge for a low signal to noise ratio and lacking of shape and texture information. Aiming at the problem of accurate eye localization and segmentation, simplified pulse coupled neural networks (SPCNN) combined with morphology method is put forward. The contributions of this work can be divided into two parts. The first contribution is that the local region of the eyes is extracted efficiently via morphology opening top-hat operator, as the region of interest for ensuring the follow-up processing without interference of the background. The second contribution is that a SPCNN model is proposed to carefully partition pixels into a corresponding cluster in iterative manner for ensuring high segmentation performance. Experiments are carried out on the near infrared images obtained by the designed acquisition system using the proposed method as well as Otsu and k-means for comparison. Experimental results show that our method achieves desired segmentation performance and has a lower misclassification error.

Because of the lenses’ limited depth of field, digital cameras are incapable of acquiring an all-in-focus image of objects at varying distances in a scene. Multi-focus image fusion technique can effectively solve this problem. Aiming at the block-based multi-focus image fusion methods, the problem that blocking-artifacts often occurs. An Adaptive block-based fusion method based on lifting undistorted-edge structural similarity (LUE-SSIM) is put forward. In this method, image quality metrics LUE-SSIM is firstly proposed, which utilizes the characteristics of human visual system (HVS) and structural similarity (SSIM) to make the metrics consistent with the human visual perception. Particle swarm optimization(PSO) algorithm which selects LUE-SSIM as the object function is used for optimizing the block size to construct the fused image. Experimental results on LIVE image database shows that LUE-SSIM outperform SSIM on Gaussian defocus blur images quality assessment. Besides, multi-focus image fusion experiment is carried out to verify our proposed image fusion method in terms of visual and quantitative evaluation. The results show that the proposed method performs better than some other block-based methods, especially in reducing the blocking-artifact of the fused image. And our method can effectively preserve the undistorted-edge details in focus region of the source images.

With the tendency of aging society, countries all over the world are dealing with the demographic change. Fall had been proven to be of the highest fatality rate among the elderly. To realize the elderly fall detection, the proposed algorithm used the hybrid feature. Based on the rate of centroid change, the algorithm adopted VEI to offer the posture feature, this combined motion feature with posture feature. The algorithm also took advantage of SIFT descriptor of VEI(V-SIFT) to show more details of behaviors with occlusion. An improved motion detection method was proposed to improve the accuracy of front-view motion detection. The experimental results on CASIA database and self-built database showed that the proposed approach has high efficiency and strong robustness which effectively improved the accuracy of fall detection.

Distributed machine vision system was applied for the detection on the cable surface defect of the cable-stayed bridge, and access to surface defects including longitudinal cracking, transverse cracking, surface erosion and scarring pit holes and other scars. In order to achieve the automatic classification of surface defects, firstly, part of the texture features, gray features and shape features on the defect image were selected as the target classification feature quantities; then the particle swarm optimization (PSO) was introduced to optimize the punitive coefficient and kernel function parameter of the support vector machine (SVM) model; and finally the objective of defects was identified with the help of the PSOSVM classifier. Recognition experiments were performed on cable surface defects, presenting a recognition rate of 96.25 percent. The results showed that PSO-SVM has high recognition rate for classification of surface defects on bridge cable.

Aiming at the block-based fusion methods for multi-focus images, such key problems like finding a suitable block size and overcoming the blocking-artifact often occurs. To address them, we present an intelligent fusion method using particle swarm optimization (PSO) algorithm, which is guided by a structural similarity (SSIM) criterion. In this method, PSO is to fast search for an appropriate block size, while the SSIM is employed as the fitness function in PSO, in order to evaluate the blocking-artifact, thus enabling this method to have a significant improvement in fusion performance. Experimental results show that the proposed method performs better than some classical methods, such as a block-based method similar with ours in fusion rule and multiresolution-based method. Besides, our method preserves more details than others in fused image.

In the research and development of multi-parameter precision detecting system, various parameters need to be measured (such as length, diameter, surface roughness and so on). We not only measure some parameters statically but also measure others dynamically and calculate others (such as volume, density) using some detected parameters. At the same time, this system need higher precision and higher measuring speed. We propose a new detecting idea for this system--detecting devices in dynamic, module and time-sharing. And we design and optimize multi-parameter high precision measurement system employing the method. This idea includes three parts: the first part is dynamical part. We can make system more stability and more continuity in high detecting speed. The second part is module part. We can settle on a solution to measure similar parameters and make system structure more reasonable and reduce error factors. The third is time-sharing part. We solve the problem how to allot time to every parameter and make every measurement part and its software tie in. The detecting idea has been employed to design and optimize several multi-parameter precision detecting systems. Now these systems are running successfully in workshop.

This paper studies near infrared technology for measuring moisture content in solid substance. After analyzing and improving traditional moisture instruments and discussing their disadvantages, we present a measuring scheme and instrument structure design with high performance and stability. The novel instrument utilizing near infrared range analysis (NIRA) technology possesses characteristics of on-line working, high accuracy, distribution and non-contact measurement.

Different methods of optical imaging of biological media based on time-resolved and low-coherence techniques are discussed, including the use of streak camera, Kerr gate, electronic holography and the low-coherence interferometry. The time-resolved techniques use ultrashort laser pulses and the photons traveling different paths can be detected selectively because they have different time of flight. The low-coherence techniques use light sources with broad spectrum and low coherence length to achieve high resolution.

This paper discusses to establish the characteristic model of particles' Brownian motion by fractal theory. Particles' Brownian motion is a kind of moving at random and particles' position oscillates with extreme irregularity. After having analyzed this process, authors present a fractal model to describe the physical process and prove theoretically the model is valid. The relationship between fractal characteristic information and particle size parameters is deduced. Combining image processing technique, it can provide a new method and technique of submicron particle sizing.