Identification of Impurities in Fresh Shrimp Using Improved Majority Scheme-Based Classifier

Liu, Zihao; Fang, Cheng; Hong, Hanmei

doi:10.1007/s12161-016-0497-3

Cited by 8 publications

(4 citation statements)

References 36 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rossi et al [ 26 ] also developed another system for the identification of fish species called FishAPP using a machine vision system coupled with ANN. Liu et al [ 27 ] used a machine vision system and improved the majority rule (IMAJ) classifier to identify impurities in fresh shrimp. They showed that IMAJ based classifier combined with parallel features are superior (91.53%) over other fusion rule-based classifiers.…”

Section: Introductionmentioning

confidence: 99%

Comparative Performance Analysis of Support Vector Machine, Random Forest, Logistic Regression and k-Nearest Neighbours in Rainbow Trout (Oncorhynchus Mykiss) Classification Using Image-Based Features

Saberioon

Císař

Labbé

et al. 2018

Sensors

View full text Add to dashboard Cite

The main aim of this study was to develop a new objective method for evaluating the impacts of different diets on the live fish skin using image-based features. In total, one-hundred and sixty rainbow trout (Oncorhynchus mykiss) were fed either a fish-meal based diet (80 fish) or a 100% plant-based diet (80 fish) and photographed using consumer-grade digital camera. Twenty-three colour features and four texture features were extracted. Four different classification methods were used to evaluate fish diets including Random forest (RF), Support vector machine (SVM), Logistic regression (LR) and k-Nearest neighbours (k-NN). The SVM with radial based kernel provided the best classifier with correct classification rate (CCR) of 82% and Kappa coefficient of 0.65. Although the both LR and RF methods were less accurate than SVM, they achieved good classification with CCR 75% and 70% respectively. The k-NN was the least accurate (40%) classification model. Overall, it can be concluded that consumer-grade digital cameras could be employed as the fast, accurate and non-invasive sensor for classifying rainbow trout based on their diets. Furthermore, these was a close association between image-based features and fish diet received during cultivation. These procedures can be used as non-invasive, accurate and precise approaches for monitoring fish status during the cultivation by evaluating diet’s effects on fish skin.

show abstract

Section: Introductionmentioning

confidence: 99%

Comparative Performance Analysis of Support Vector Machine, Random Forest, Logistic Regression and k-Nearest Neighbours in Rainbow Trout (Oncorhynchus Mykiss) Classification Using Image-Based Features

Saberioon

Císař

Labbé

et al. 2018

Sensors

View full text Add to dashboard Cite

show abstract

“…Although the proposed method is costly, however, the segmentation time is fully competent to perform signals transmission from industrial computer to single-chip in the on-line quality detection and classification system. That is because it has a short distance for object transportation from the stage of capturing image to the stage of actuator [30,31]. In recent years, with the development of Fig.…”

Section: Resultsmentioning

confidence: 99%

High‐order non‐linear split paths construction algorithm for clustered axisymmetric agriculture products

Liu

2021

Food Measure

Self Cite

View full text Add to dashboard Cite

A special high-order touched pattern was found when performing quality detection, counting and classification for clustered axisymmetric agriculture products. A modified strategy of single-circle segmentation algorithm based on symmetrization and multiple-circle fitting was proposed to split this pattern. Firstly, to obtain region of interest from original image, vision saliency detection algorithm was used to extract contour of object image. Then, a circular mask based central point extraction algorithm was applied into the contour, get a set of points. After that, pairwise of optimization concavity points were found and filtered according to concaveness feature and four restrict acceptance conditions. Furthermore, proper linear segmentation lines were formed by connecting the pairwise of optimization concavity points. Low-order non-linear segmentation lines can also be built by finding a new point based on single-circle fitting from our previous studies. To construct the high-order non-linear segmentation lines especially existed in axisymmetric agriculture products, major new points were found and built by using the unique symmetry and geometry properties of the research target. Finally, linear, low-order non-linear and highorder non-linear segmentation lines were integrated in an original image to achieve final segmentation results. Experimental results showed that average segmentation accuracy with 95.46%, 91.95 and 92.63% were respectively achieved across three clustered image datasets-peanut, kidney bean and green soy bean. The segmentation results confirm that this method has capacity to segment axisymmetric agricultural products successfully.

show abstract

“…In [20], a joint feature extraction method that is associated with a single classifier is addressed, which has a low classification rate. This work mainly focused on classification of impurities presented in the fresh shrimps.…”

Section: Related Workmentioning

confidence: 99%

Transfer Learning-based Optimal Feature Selection with DLCNN for Shrimp Recognition and Classification

2022

IJIES

View full text Add to dashboard Cite

Shrimp is a main international food item with a significant economic value, as well as one of the most vital animal protein sources. The variety of shrimps can be found in aquaculture. Thus, it is necessary to categorize each shrimp. The conventional machine learning approaches are failed to classify the multiple classes of shrimp, which causes large economic losses in the shrimp farming industry. Therefore, this article proposes a hybrid mechanism for shrimp recognition and classification (SRC), which is named as transfer learning-based optimal feature selection (TLOFS) with deep learning convolutional neural network (DLCNN). Initially, transfer learning based AlexNet is used to extract the features from the shrimp samples. Then, machine learning based iterative random forest algorithm (IRFA) is utilized to select the optimal features from the AlexNet extracted features, which can also identify the relationship between various shrimp classes. Finally, DLCNN is trained and tested with the optimal features and classifies the various shrimp categories, hereafter the proposed hybrid model is named as TLOFS with DLCNN. Obtained simulations discloses the effectiveness of proposed TLOFS with DLCNN model with 99.98% of accuracy, and 99.97% of F1-score as compared to state-of-art SRC approaches.

show abstract

Identification of Impurities in Fresh Shrimp Using Improved Majority Scheme-Based Classifier

Cited by 8 publications

References 36 publications

Comparative Performance Analysis of Support Vector Machine, Random Forest, Logistic Regression and k-Nearest Neighbours in Rainbow Trout (Oncorhynchus Mykiss) Classification Using Image-Based Features

Comparative Performance Analysis of Support Vector Machine, Random Forest, Logistic Regression and k-Nearest Neighbours in Rainbow Trout (Oncorhynchus Mykiss) Classification Using Image-Based Features

High‐order non‐linear split paths construction algorithm for clustered axisymmetric agriculture products

Transfer Learning-based Optimal Feature Selection with DLCNN for Shrimp Recognition and Classification

Contact Info

Product

Resources

About