Search

Digital image of pipeline weld is an important basis for the reliability management of pipeline welds. However, the error rate of artificial discrimination is high. In order to increase the defect identification accuracy of digital image of pipeline weld, we adopted several methods (e.g. multiple edge detection, detection channel and threshold segmentation) to carry out image processing on the image defects of pipeline welds. Then, a defect characteristic database on the digital images of pipeline welds was constructed, including grayscale difference, equivalent area (S/C), circularity, entropy, correlation and other parameters. Furthermore, a multi-classifier construction (SVM) model was established. Thus, the classification and evaluation on the defects in the digital images of pipeline welds were realized. Finally, an automatic defect identification software for digital image of pipeline weld was developed and verified on site. And the following research results were obtained. First, after image processing, the edge detection results obtained by Canny and other algorithms are satisfactory when there is no noise. In the case of noise, however, pseudo-edge emerges in the detection results. In this case, the automatic threshold selection method shall be adopted to detect the image edge to obtain the rational threshold. Second, there are 14 parameters in the defect characteristic database, including shape characteristic, lamination characteristic and image length pixel. Third, by virtue of the SVM classification model, the shape characteristics of each type of defect can be clarified, and the defect characteristics can be identified, such as crack, slag inclusion, air hole, incomplete penetration, non-fusion and strip. Based on field application, the following results were obtained. First, this automatic defect identification technology is applicable to quality identification and evaluation of various defects in pipeline welds. Second, its identification accuracy is higher than 90%. Third, by virtue of this technology, automatic defect identification and evaluation of digital image of pipeline weld is realized. In conclusion, these research results help to ensure the safe operation of pipelines. Keywords: Pipeline weld, Ray film, Digital image, Defect database, SVM classification model, Defect identification, Automatic identification, Software development

At present, mobile robots equipped with a single sensor in an indoor environment suffer from insufficient mapping accuracy and limited scanning range. Therefore, this paper proposes a method for data fusion mapping with three sensors, single-line lidar, depth camera and IMU. First, the depth data collected by the depth camera is reduced by dimensionality processing to make it two-dimensional, and then the environmental feature data scanned in the lidar, the pseudo laser data of the depth camera, and the pose information collected and calculated by the IMU are passed through Kalman. Filtering performs data fusion, that is, compensating the position and attitude errors caused by the lidar measurement. In the mapping phase, under the open source SLAM algorithm Gmapping, the two-dimensional local raster maps generated by the lidar that fuse multi-source data and the depth camera are merged with the local map based on the Bayes rule. Experiments show that the global map obtained by this method contains richer environmental information than a single sensor, which improves the accuracy of map construction and is beneficial to subsequent navigation obstacle avoidance research.

The SURF algorithm is widely used because of its high registration accuracy. However, because the algorithm detects the feature points of the entire image, it cannot meet the characteristics of single texture, similarity and density of fabric images. At the same time, because of the long search time and low computational efficiency of image registration using KNN method for feature detection and matching, this paper studies the image registration algorithm based on KD tree search and SURF features. The input of the algorithm is to register the set of feature points in the image, and generate a feature description vector, and then establish the KD tree index of the feature description vector. When performing image registration, the ratio method is used as the criterion for matching feature points. Each feature point in the registered image is regarded as a query point, and the first two feature points closest to the query point are found in the KD tree. Finally, the MASC algorithm combining geometric motion parameter estimation and interpolation is used to purify the matching points. The experimental results show that the KD tree-based SURF algorithm, which combines the MASC algorithm to eliminate mismatched point pairs, is better than other algorithms in terms of speed and accuracy in feature detection.

The sensing part of the electronic current transformer is composed of an integrating circuit, an amplifying circuit, an A/D conversion circuit and the like, and these lines are completely electronic circuits. Therefore, the high-voltage side of the electronic current transformer must have a stable power supply to these electronic components line. It is a difficult and important point in the circuit power supply problem of electronic current transformer designers. A stable power supply solution is an important condition for ensuring stable operation of the electronic current transformer. In this paper, based on the high-voltage energy extraction of the energizing coil and the lithium-ion battery combined power supply scheme, according to the established simulation model of the energy-carrying coil, the optimal design and simulation of the optimal number of energizing coils are carried out. The feasibility and effectiveness of the scheme are verified by simulation experiments on the energy-carrying coil and the power supply system.

The Principal Component Analysis (PCA) algorithm is widely used in the field of face recognition because of its high recognition rate and simplicity. The PCA algorithm is based on the principle of Karhunen-Loeve Transformation, because the PCA algorithm is sensitive to outliers, it is improved on the basis of PCA algorithm, combined with Linear Discriminant Analysis (LDA) algorithm, the PCA-LDA face recognition method is proposed. This method obtains the feature space of training sample set by PCA algorithm, On this basis, the LDA algorithm is executed to obtain the feature space of fusion. The PCA is then fused with the LDA's feature space to obtain the new feature space that combines the two. Finally, the face projected in the new feature space is trained and recognized. The experimental results show that the face recognition algorithm proposed in this paper has a higher recognition rate than the traditional PCA algorithm.

Aiming at the problem of bad working environment of bulldozer in landfill, this paper presents a target detection method suitable for autonomous working bulldozer. Firstly, the working environment information was obtained by the binocular camera, the maximum inter-class variance method was used to remove the sky background by binarization of the image, and the color component of the ground was extracted from the HSV color space and converted into a binarization image. The absolute value of the difference between the two binarization images was the approximate area of the working target. Then the binocular camera was calibrated and the feature points were detected and matched in the target area. The parallax was obtained by triangulation principle to obtain the depth information of the target. The experimental results show that the method can correctly extract the depth information of the target area and target, and the accuracy is high, which can meet the requirements of the bulldozer in the landfill for target detection.

For the vibration problem of the squirrel-cage cleaning machine roller under normal load, modal analysis for roller was carried out to reduce the occurrence of resonance.The paper used 3D design software Solidworks to set up 3D solid model for squirrel-cage cleaning machine roller.The finite element software Ansys workbench was used for modal analysis. The results showed that natural frequency of the first two order was 7.6931Hz and 8.3117Hz.It was greater than the vibration frequency of roller at 0.7766Hz and less than the vibration frequency of motor at 23.33Hz, thus the resonance would not occur.

Cleaning is an important link in the production of flour milling. In this paper, a spray-type cleaning device is designed with sweet potato as the object of study, and three spray piping is installed in the device, and the sweet potato epidermis is cleaned by high pressure water jet. In the selection of nozzles, in order to ensure good cleaning effect, the nozzle aperture and spray angle must be reasonable, the outlet pressure value reached 3.0kg to produce atomization effect, nozzle jet characteristics and atomization effect is an important indicator of the cleaning device. Pump body selection is also the key to the whole device, according to the pump characteristic curve selection of the appropriate pump type, both to ensure water pressure and save water consumption. The whole plant can also play a good cleaning effect on the same type of root vegetable, and has a strong popularization.

This paper analyzes the current status of existing sweet potato powder processing equipment, optimizes existing processing equipment, and designs automated processing equipment based on PLC for sweet potato cleaning, pulping, mixing and milling. The S7-200 series from Siemens is used as the main control unit of this system. The liquid level sensor and temperature controller are added to study the hardware configuration, input/output point allocation and software design method of the control system. The equipment can save time, labor and high efficiency in the processing and processing of sweet potato powder, which is much more efficient than the traditional manual processing method, and also shortens the capital turnover time. The realization of automatic processing equipment for cleaning, pulping, mixing and milling is the basis for intelligent and automated production of sweet potato milling equipment, with high practical value and application prospects.

In order to realize the identification and positioning of the working object when the excavator works independently, it is necessary to calibrate the binocular camera to determine that the binocular is the internal and external parameters of the camera. This paper focuses on the calibration of binocular camera, and determines the transformation relationship among world coordinate system, camera coordinate system and image coordinate system based on binocular stereo vision ranging principle. The checkerboard images that meet the calibration requirements are selected by using the visual toolbox of MATLAB. The camera calibration process is carried out to determine the camera parameters to be determined. The calibration parameters are analyzed and compared to preliminarily verify the accuracy of the calibration parameters. Finally, the error source is analyzed in QT5.14.2 programming environment combined with OpenCV3.4.10, the calibration method meets the application requirements.

With the continuous improvement of people's living standard, people's material needs have been met. Nowadays, some of the unhealthy living conditions have led to outbreaks of disease, and have also caused some damage to people's healthy life. For example, alcohol consumption has a greater impact on people's health. Based on the prevalence of alcoholism in current society, the postoperative nursing and related effects were further analysed in this paper. The purpose of this study is to provide some theoretical basis and scientific support for the improvement of our country's physical health.

A new method is proposed to solve the problem of image restoration of high resolution TDICCD camera due to satellite vibrations, which considers image blur and irregular sampling geometric quality degradation simultaneously. Firstly, the image quality degradation process is analyzed according to imaging characteristics of TDICCD camera, owing to image motions during TDICCD integration time induced by satellite vibrations. In addition, the vibration simulation model is established, and the irregular sampling degradation process of geometric quality is mathematically modeled using bicubic Hermite interpolation. Subsequently, a full image degradation model is developed combined with blurred and noisy degradation process. On this basis, a new method of image restoration is presented, which can implement not only deblurring but also irregular to regular sampling. Finally, the method is verified using real remote sensing images, and compared with the recent restoration methods. Experimental results indicate that the proposed method performs well, and the Structural Similarity between the restored and ideal images are greater than 0.9 in the case of seriously blurred, irregularly sampled and noisy images. The proposed method can be applied to restore high resolution on-orbit satellite images effectively.

Aiming to the difficulty of micro pressure detection and the difficulty of micro flow control in the testing process of hydrocephalus shunt, the principle of the shunt performance detection was analyzed.In this study, the author analyzed the principle of several items of shunt performance detection,and used advanced micro pressure sensor and micro flow peristaltic pump to overcome the micro pressure detection and micro flow control technology.At the same time,This study also puted many common experimental projects integrated, and successfully developed the automatic detection system for a shunt performance detection function, to achieve a test with high precision, high efficiency and automation.

We propose an in-orbit modulation transfer function (MTF) statistical estimation algorithm based on natural scene, called SeMTF. The algorithm can estimate the in-orbit MTF of a sensor from an image without specialized targets. First, the power spectrum of a satellite image is analyzed, then a two-dimensional (2-D) fractal Brownian motion model is adopted to represent the natural scene. The in-orbit MTF is modeled by a parametric exponential function. Subsequently, the statistical model of satellite imaging is established. Second, the model is solved by the improved profile-likelihood function method. In order to handle the nuisance parameter in the profile-likelihood function, we divided the estimation problem into two minimization problems for the parameters of the MTF model and nuisance parameters, respectively. By alternating the two iterative minimizations, the result will converge to the optimal MTF parameters. Then the SeMTF algorithm is proposed. Finally, the algorithm is tested using real satellite images. Experimental results indicate that the estimation of MTF is highly accurate.

Driver's fatigue and spirit distraction has become one of the most important causes for traffic accidents. The detection of drivers mouth state could offer a useful distinguish criterion for judge whether the driver is fatigue or spirit distraction. Based on our research foundation, a new method with certain environmental robustness for driver's face detection and location, mouth detection and location is proposed based on color space. This method could be establishing a good step for further research on the capture of driver's mouth state.

The weight in TOPSIS approach (technique for order preference by similarity ideal solution - TOPSIS ) is given by experts or decision makers. The value of weight would be influenced by experts’ subjective judgments. A slight difference in value of weight may result in diversity of order of alternatives. In this paper, a Bayesian method for decision of weight for MADM model with interval data is introduced. The value of weight is decided by prior information (other experts’ knowledge, or numerical simulation etc.) and experts’ knowledge (or decision makers’ experience/ preference). This method effectively takes advantage of experts’ knowledge and avoids the problem with experts’ subjectivity. An illustrative example is showed to explore the applications of proposed method. The method is valuable for field of multi-attribute decision- making with interval data.

One of hotspot of the existing approaches of multi-attribute decision making model (MADM) is determination of weights. This paper presents a novel kind of decision of weight of multi-attribute decision-making model based on Bayesian networks. Critical technique, dasiaBayesian networkspsila is used to determine value of weights, which combined prior information (other expertspsila knowledge, or numerical simulation etc.) with expertspsila knowledge. This method effectively takes advantage of expertspsila knowledge and avoids the problem with expertspsila subjectivity. An illustrative example is showed to explore the implications of proposed method. The model is valuable for field of multi-attribute decision-making.