The pick is a key component of shearer,and its performance directly affects the structural strength and service life of the shearer.Wear is the main form of pick’s failure of shearer.The study of wear is very important for the timely replacement and improvement of picks,and it is also of great significance to improve the production efficiency and reduce the production cost of shearers.Due to the limitation of measurement methods,the current measurement method of shearer pick is complex,inefficient and with insufficient accuracy.Shape from focus technology can realize irregular surface topography restoration and height measurement.The method is simple,easy to operate and low in cost.It provides a practical measurement scheme for the analysis of the wear mechanism of the pick and calculation of the wear rate.In this paper,in order to improve the reconstruction accuracy of shape from focus,in view of the shortcomings of the existing shape from focus,starting with its key technologies and the key influencing factors of its performance,an improved algorithm for shape from focus techniques was studied.In the aspect of image preprocessing,the max-min operator filtering method is proposed.In the focus measure algorithm,adaptive window selection method in shape from focus and shape from focus based on image regional pixel reconstruction are proposed.Finally,the improved algorithm of shape from focus was applied to the wear measurement of shearer pick,and a pick image acquisition test bench was set up.The three-dimensional surface topography restoration and volume calculation of the pick were completed by using the collected pick image and the principle of shape from focus,and the accuracy of the pick measurement can be verified through experiments.In view of the salt and pepper noise in the image,the conventional erroneous image filtering method is abandoned and the max-min operator image filtering method is proposed.The max-min operator is designed.This operator can detect the noise points in the image efficiently and the new gray value is given to the noise pixel only by using the nonlinear median filter method.This method not only can effectively remove the image noise,but also can maintain the original information of the image to the greatest extent,which provides the basis for the accurate calculation of the following focused evaluation value.An adaptive window selection method for shape from focus is proposed.The evaluation window is an important factor affecting the accuracy of evaluation function value calculation.Too small evaluation window will be sensitive to the image noise and increase the probability of error value.As the evaluation window increases,the error value will be significantly reduced,but the recovery surface will be overly smooth and lose surface details.Normally choosing 5×5 is the best choice.The adaptive window selection method determines the evaluation window size of the pixel according to the gray correlation of the neighboring pixels.Experiments have shown that the adaptive window can not only reduce the error value but also the corresponding recovery morphology is closest to the original surface,and the comprehensive effect is improved obviously.A shape from focus algorithm based on image region pixel reconstruction is proposed.In the area of 2 x 2 pixels,taking the average gray value of pixels in this area as its gray value and performing the regional pixel reconstruction on the image sequence.According to the gray difference of the adjacent pixels in the reconstructed image,the focus measure is designed to determine the focus level of the region.The experimental results show that compared with the traditional focus measure method,both the noisy image sequence and the no-noise image sequence,the focus measure algorithm designed in this paper has better accuracy.It is feasible and effective.Based on the shape from focus technology,the above-mentioned improved focus topography restoration algorithm was used to complete the 3D surface topography recovery and volume measurement scheme are designed.According to the requirement of the scheme,the image acquisition test bed was set up to realize the collection of the image sequence of the pick,and the collected image was used to recover the complete surface appearance of the pick tip,and the volume was measured using the surface depth information.Furthermore,a verification experiment was conducted to prove the effectiveness of the experimental scheme and the accuracy that the pick measurement experiment can achieve.The result proves that this measurement method can make the volumetric accuracy of the measurement reach more than 97%. |