TY - GEN
T1 - Convolutional Neural Network-Based Quality of Fruit Detection System Using Texture and Shape Analysis
AU - Mishra, Om
AU - Parashar, Deepak
AU - Harikrishanan,
AU - Gaikwad, Abhinav
AU - Gagare, Anubhav
AU - Darade, Sneha
AU - Bhalla, Siddhant
AU - Pandey, Gaurav
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.
PY - 2023
Y1 - 2023
N2 - This research describes an excellent method for detecting fruits’ quality using convolutional neural networks. Fruit grading is done by inspections, experiences, and observation. To rate the quality of fruits, the proposed system employs machine learning techniques. Shape and color-based analysis methods are used to grade two-dimensional fruit depictions. Different fruit photos may have identical color, size, and shape qualities. As a result, utilizing color or form property analysis methods to identify and differentiate fruit photos is ineffective. As a result, we combined a size, shape, and color-based method with a CNN to improve accuracy and precision of fruit quality recognition. The advisable system begins the process by selecting the fruit images. The image is then sent to the rectification stage, where fruit sample properties are retrieved. Subsequently, fruit images are trained and tested using a CNN. The convolutional neural network is used in this proposed paper to abstract out colors, size, and shape of the fruits and results achieved with a combination of these features are quite promising.
AB - This research describes an excellent method for detecting fruits’ quality using convolutional neural networks. Fruit grading is done by inspections, experiences, and observation. To rate the quality of fruits, the proposed system employs machine learning techniques. Shape and color-based analysis methods are used to grade two-dimensional fruit depictions. Different fruit photos may have identical color, size, and shape qualities. As a result, utilizing color or form property analysis methods to identify and differentiate fruit photos is ineffective. As a result, we combined a size, shape, and color-based method with a CNN to improve accuracy and precision of fruit quality recognition. The advisable system begins the process by selecting the fruit images. The image is then sent to the rectification stage, where fruit sample properties are retrieved. Subsequently, fruit images are trained and tested using a CNN. The convolutional neural network is used in this proposed paper to abstract out colors, size, and shape of the fruits and results achieved with a combination of these features are quite promising.
UR - https://www.scopus.com/pages/publications/85174507300
UR - https://www.scopus.com/pages/publications/85174507300#tab=citedBy
U2 - 10.1007/978-981-99-5085-0_36
DO - 10.1007/978-981-99-5085-0_36
M3 - Conference contribution
AN - SCOPUS:85174507300
SN - 9789819950843
T3 - Lecture Notes in Networks and Systems
SP - 379
EP - 389
BT - Advances in IoT and Security with Computational Intelligence - Proceedings of ICAISA 2023
A2 - Mishra, Anurag
A2 - Gupta, Deepak
A2 - Chetty, Girija
PB - Springer Science and Business Media Deutschland GmbH
T2 - International Conference on Advances in IoT, Security with AI, ICAISA 2023
Y2 - 24 March 2023 through 25 March 2023
ER -