论文部分内容阅读
为了提高模具表面缺陷检测的质量,采用二进制粒子群算法。首先建立测量模型,包括光学景深模型、成像系统景深模型、系统成像几何模型标定。接着对粒子位置的每一维分量被限制为1或0,对应模具表面检测过程中的有缺陷或无缺陷判决;然后利用sigmoid函数将粒子的速度转换到区间[0,1]上,并且修正粒子的位置;最后在二维平面内建立有限元模型和判决准则。实验仿真给出了各个方向的数据优化效果和不同算法空间数据迭代效果,相比其它算法新的算法对模具表面缺陷检测的过检率、正检率、漏检率指标较好。
In order to improve the quality of mold surface defect detection, binary particle swarm optimization is adopted. First, the measurement model is established, including the optical depth of field model, the depth of field imaging system model, and the system imaging geometric model calibration. And then the particle position of each dimension is limited to 1 or 0, corresponding to the mold surface detection process of defect or defect-free decision; and then use the sigmoid function to the particle velocity is converted to the interval [0,1], and the amendment Finally, the finite element model and decision criterion are established in two-dimensional plane. The experimental simulation shows the data optimization effect in different directions and the spatial data iteration effect of different algorithms. Compared with the other algorithms, the new algorithm has better detection rate, positive detection rate and missed detection rate of the mold surface defect detection.