Description

CLASSIFICATION OF CHICKEN MEAT FRESHNESS USING CONVOLUTIONAL NEURAL NETWORK ALGORITHMS

ABSTRACT

Broiler chicken meat is one of the most widely consumed meat types in Indonesia, this high level of consumption makes a lot of consumer demand in the market. However, there was a found seller who sells broiler chicken meat that is rotten. In this study, we develop chicken meat freshness identification using a convolutional neural network algorithm. This study used the image dataset of broiler chicken breasts. There are two categories of chicken meat used in the study, namely, fresh and rotten. The meat images were acquired by using a smartphone camera. For the process of cropping chicken meat images, we use thresholding with the Otsu method and conversion of RGB images to binary images to select the area of RGB images before cropping the images. The chicken meat images were cropped into three sizes and then used as a dataset in the study. The chicken meat image dataset was trained using a simple architecture that was self-made. From the experiment results, we can conclude that using Ayam6Net architecture with dataset 400 × 400 pixels has a better accuracy result compared with other architectures and other sizes image datasets.

INTRODUCTION

Poultry meat is one of the most consumed types of meat in Indonesia with a percentage level of 70% and almost 85% of poultry meat consumed is broiler chicken. Besides affordable prices, broiler chicken meat has a high protein content compared to pork, beef, and lamb. The high level of consumption of chicken meat is accompanied by a lot of consumer demand on the market. However, in several cases, such as the increasing price of chicken meat, the chicken meat in the market did not sell well, causing sellers to commit fraudulent acts by selling rotten chicken meat at this time as a case that appeared in the media. One of the efforts to identify rotten chicken meat is by visual based on the color and texture characteristics of the chicken meat. Digital image processing is a method that can be used to identify the quality of chicken meat based on the color characteristics and texture characteristics of the meat. Several studies on the classification of meat-based on images have been carried out, such as classification of rotten meat based on readings from gas sensors and image processing, classification of red meat from hyperspectral images, and identification of freshness of chicken based on color and texture features. In this study, the feature extraction method, as well as the classification method for chicken meat images, used a Convolutional Neural Network (CNN) algorithm.

PROPOSED SYSTEM

In this paper, the method used for image classification of chicken meat consists of 3 main steps. There are image acquisition, image pre-processing, and image classification using CNN. At the image acquisition step, samples of chicken meat are photographed using a smartphone camera. In the second step, image pre-processing aims to increase the features of the chicken meat image to be processed on CNN. The last step is the process of classifying the image of chicken meat.

BLOCK DIAGRAM

CONCLUSION

In this study, we proposed an identification of chicken meat freshness using a Convolutional Neural Network algorithm. The freshness level of chicken meat is divided into two categories, fresh (6-8 hours after slaughtered) and rotten (21- 23 hours after slaughtered). In this experiment, we use a smartphone camera for the acquisition process of chicken meat images. From the experiment results, we can conclude that the best result was achieved using this architecture with the dataset 400 × 400 pixel where the training and validation accuracy has over 95% and the test result accuracy has a 92.9%

For more Image Processing projects ,Click the link below

https://www.pantechsolutions.net/image-processing-projects