摘要: |
探讨了判别分析确定最佳阈值的快速算法,以解决经典算法处理速度过慢,不能满足计算机视觉对农作物及农产品进行实时检测和分级要求的问题。利用图像的灰度直方图数据进行类分布参数的统计,从而避免重复对整修图像数据进行统计计算,以提高处理速度。结果表明,快速算法可使处理时间减小94.7%~97.5%。 |
关键词: 判别分析法 阈值 图像处理 直方图 |
DOI: |
分类号: |
基金项目:陕西省自然科学基金资助项目(4130103) |
|
Fast discriminant analysis algorithm for the best threshold |
|
Abstract: |
The paper discusses a fast algorithm to improve the speed of the origina discriminant analysis algorithm for the best threshold.It can solve the problem that the original algorithm requires much more time with low speed while the image size increases and can’t satisfy the need of real time detection and classification of agricultural products with computer vision.The result indicates the fast algorithm can decrease the processing time by 94.7%-97.5%. |
Key words: discriminant analysis threshold image processing histogram |