A Fast Beef Marbling Segmentation Algorithm Based on Image Resampling

Bin Pang, Xiao Sun, Xin Sun, Kunjie Chen

Abstract


With the miniaturization and portability of online detection and grading equipment, traditional PC is being replaced with ARM or DSP embedded systems in beef quality grading industry. As the low basic frequency of embedded system, the traditional beef marbling segmentation method can not meet requirements of real-time performance. The fast segmentation algorithm of beef marbling based on image resampling is put forward aiming the disadvantages that the traditional method is time-consuming and does not apply to embedded systems. First, the entropies of the original image and resampling image were calculated according to the entropy principle to determine the image resampling rate based on entropy constraint according to the changes of relative information entropy of resampling image. Then fuzzy c-mean (FCM) cluster segmentation was conducted on the resampling image to calculate the beef image segmentation threshold. Finally, beef marbling area is segmented via morphological and logic operations on a series of images. The experimental results show that this proposed algorithm took 0.57s on average in beef marbling image segmentation under the constraints that the loss rate of relative information entropy ranged between 0.5-1.0%, which is only 6.43% of that of the traditional FCM cluster segmentation algorithm, indicating significantly augmented efficiency of segmentation.

 

DOI : http://dx.doi.org/10.11591/telkomnika.v12i5.4294


Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics