Comparison of Different Feature Extraction Techniques in Biospeckle Images for Nondestructive Assessment of Apple firmness

Publish Year: 1394
نوع سند: مقاله کنفرانسی
زبان: English
View: 314

This Paper With 5 Page And PDF Format Ready To Download

  • Certificate
  • من نویسنده این مقاله هستم

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این Paper:

شناسه ملی سند علمی:

ICTINDT03_077

تاریخ نمایه سازی: 18 مرداد 1398

Abstract:

Assessment of firmness as a vital quality attribute of apple fruits has been more popular over the past years. Common methods for determination of this attribute are destructive and cannot be applied in sorting lines. Therefore, there is need to develop a nondestructive, simple, fast, and low-cost technique for assessment of apple firmness. Biospeckle imaging as a new nondestructive technique has been recently introduced in Biosystems engineering applications. This technique combined with image feature extraction methods have been recently noticed for nondestructive quality assessment of food and agricultural products. This research aims to compare different feature extraction techniques in biospeckle images of apple for nondestructive assessment of fruit firmness. To this end, biospeckle images at two wavelengths of 680nm and 780nm were acquired using a CCD camera. After creating time history of speckle patterns (THSP), some features including inertia moment (IM), the absolute value of differences (AVD), wavelet, and texture features were extracted from each THSP image to assess the fruit firmness. The highest Pearson’s correlation coefficients between apple firmness measured destructively and the extracted features from THSP images at the wavelengths of 680 and 780 were obtained for entropy based on texture features. It was also noted that some texture and wavelet features are better at description of biospeckle activity compared with usual biospeckle features (IM and AVD) to assess the fruit firmness, nondestructively

Authors

Bahareh Jamshidi

Academic Member, Agricultural Engineering Research Institute, Agricultural Research Education and Extension Organization (AREEO), Karaj, Iran;

Arman Arefi

PhD student, Urmia University, Urmia, Iran;