کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
4508353 | 1321588 | 2016 | 6 صفحه PDF | دانلود رایگان |
عنوان انگلیسی مقاله ISI
Classification of fresh and spoiled Japanese dace (Tribolodon hakonensis) fish using ultraviolet-visible spectra of eye fluid with multivariate analysis
دانلود مقاله + سفارش ترجمه
دانلود مقاله ISI انگلیسی
رایگان برای ایرانیان
کلمات کلیدی
موضوعات مرتبط
علوم زیستی و بیوفناوری
علوم کشاورزی و بیولوژیک
علوم زراعت و اصلاح نباتات
پیش نمایش صفحه اول مقاله

چکیده انگلیسی
Ultraviolet-visible (UV-VIS) spectral properties of eye fluid were used to classify Japanese dace fish into fresh or spoiled groups by support vector machine (SVM), linear discriminant analysis (LDA) based on principal component analysis (PCA) scores and using a soft independent modeling of class analogy (SIMCA) classification technique. These models were then evaluated in terms of their sensitivity, specificity and overall accuracy. The UV-VIS absorbance spectra (250-600Â nm) of eye fluid from 168 fishes and the K value of fish flesh were measured at 3Â h intervals for 36Â h. In the SVM model, the sensitivity and specificity for fresh fish was 100%, whereas in the LDA and SIMCA models it was 100% and 90%, and 80% and 90% respectively. For the spoiled group fish the sensitivity and specificity result was 100% for the SVM model, while in the LDA and SIMCA models it was 90% and 100%, and 90% and 80% respectively. The overall classification accuracy was 100%, 93% and 87% for the SVM, LDA and SIMCA models respectively. These results, particularly for the SVM model, indicate that the spectral absorbance of fish eye fluid in the UV-VIS region can accurately classify fish into fresh and spoiled groups.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Engineering in Agriculture, Environment and Food - Volume 9, Issue 1, January 2016, Pages 64-69
Journal: Engineering in Agriculture, Environment and Food - Volume 9, Issue 1, January 2016, Pages 64-69
نویسندگان
Anisur Rahman, Naoshi Kondo, Yuichi Ogawa, Tetsuhito Suzuki, Yuri Shirataki, Yumi Wakita,