【摘要】 提出了一种基于Gabor变换和灰度梯度共生矩阵(CMGG)的超声无损测温方法。通过高强度聚焦超声(HIFU)对新鲜离体猪肉组织进行辐照,实时获取辐照前后的超声图像,对其做数字减影和Gabor滤波,提取归一化后灰度梯度共生矩阵的二阶参数梯度平均、混合熵和逆差矩作为温度的表征参数。实验结果表明:Gabor滤波后的灰度-梯度共生矩阵的二阶参数梯度平均、混合熵和逆差矩与温度的相关系数分别达到0.980 4、0.985 4和-0.964 4,优于传统的灰度均值法和实值Gabor系数法。Gabor滤波后的灰度-梯度共生矩阵的二阶参数混合熵在37℃~55℃能较好地反映温度变化信息,梯度平均则在55℃~85℃有更好的测温效果。更多还原

【Abstract】 A method of ultrasonic noninvasive temperature measurement based on Gabor transform and gray level gradient co-occurrence matrix( CMGG) is proposed. Irradiating fresh pork tissue by high intensity focused ultrasound( HIFU),ultrasound images before and after irradiation are obtained in real-time. And these images are processed by digital subtraction and Gabor filtering. Then three paramters are used as the characterization parameters of temperature,which are gradient average of the second order parameter,mixed entropy and negative moment of normalized gray gradient co-occurrence matrix. Experimental results show that gradient average of the second order parameter,mixed entropy,deficit moment and the temperature correlation coefficient of gray gradient co-occurrence matrix after Gabor filtering respectively reaches 0.980 4,0.985 4 and-0.964 4,better than traditional gray scale average method and real Gabor coefficient method. After Gabor filtering,mixed entropy of the second order paramete of the gray gradient co-occurrence matrix can reflect the change information of temperature from 37 ℃ to 55 ℃. For gradient average,there is a better measurement effect of temperature from 55 ℃ to 85 ℃.更多还原