Back
Submited Data
aaa
vvv
Comparison Data
Systematic Literature Review on the Application of Convolutional Neural Networks for Rambutan Fruit Classification: Advances, Challenges, and Future Directions
Rambutan (Nephelium lappaceum L.) is a tropical fruit widely cultivated in Southeast Asia, including Indonesia. Manual classification of rambutan types and ripeness levels remains a challenge due to the high subjectivity and time-intensive nature of the process, particularly in large-scale agricultural operations. Convolutional Neural Network (CNN), a deep learning approach, offers significant potential in automating and improving the accuracy of fruit classification tasks by extracting complex visual features such as color and texture. This study employs a Systematic Literature Review (SLR) to evaluate the application of CNN in rambutan classification. Relevant research from 2019 to 2024 was analyzed to identify trends, accuracy levels, and challenges in utilizing CNN for this purpose. Results demonstrate that CNN achieves superior accuracy (>90%) compared to traditional methods like K-Nearest Neighbor (KNN). However, limitations include restricted dataset diversity and insufficient testing under real-world conditions. Recommendations for future research emphasize the need for larger, more diverse datasets and integration of additional media, such as spectral data and video, to enhance model robustness
Stemming
Submited Data:
aaa vvv Comparison Data:
systematic literature review on the application of convolutional neural networks for rambutan fruit classification advances challenges and future directions rambutan nephelium lappaceum l is a tropical fruit widely cultivated in southeast asia including indonesia manual classification of rambutan types and ripeness levels remains a challenge due to the high subjectivity and time intensive nature of the process particularly in large scale agricultural operations convolutional neural network cnn a deep learning approach offers significant potential in automating and improving the accuracy of fruit classification tasks by extracting complex visual features such as color and texture this study employs a systematic literature review slr to evaluate the application of cnn in rambutan classification relevant research from 2019 to 2024 was analyzed to identify trends accuracy levels and challenges in utilizing cnn for this purpose results demonstrate that cnn achieves superior accuracy 90 compared to traditional methods like k nearest neighbor knn however limitations include restricted dataset diversity and insufficient testing under real world conditions recommendations for future research emphasize the need for larger more diverse datasets and integration of additional media such as spectral data and video to enhance model robustness
Jaro Winkler
| Word | TW | HV | Detail |
|---|---|---|---|
| 1. aaa | 130 words | 0.8 | systematic : 0.48, literature : 0.48, review : 0, on : 0, the : 0, application : 0.65, of : 0, convolutional : 0, neural : 0.5, networks : 0, for : 0, rambutan : 0.49, fruit : 0, classification : 0.47, advances : 0.67, challenges : 0.48, and : 0.6, future : 0, directions : 0, nephelium : 0, lappaceum : 0.63, l : 0, is : 0, a : 0.8, tropical : 0, widely : 0, cultivated : 0.48, in : 0, southeast : 0, asia : 0.75, including : 0, indonesia : 0, manual : 0.67, types : 0, ripeness : 0, levels : 0, remains : 0.49, challenge : 0.48, due : 0, to : 0, high : 0, subjectivity : 0, time : 0, intensive : 0, nature : 0.5, process : 0, particularly : 0.47, large : 0.51, scale : 0.51, agricultural : 0.52, operations : 0.48, network : 0, cnn : 0, deep : 0, learning : 0.49, approach : 0.67, offers : 0, significant : 0, potential : 0, automating : 0.66, improving : 0, accuracy : 0.67, tasks : 0.51, by : 0, extracting : 0.48, complex : 0, visual : 0.5, features : 0.49, such : 0, as : 0.65, color : 0, texture : 0, this : 0, study : 0, employs : 0, slr : 0, evaluate : 0.64, relevant : 0.49, research : 0.49, from : 0, 2019 : 0, 2024 : 0, was : 0.56, analyzed : 0.67, identify : 0, trends : 0, utilizing : 0, purpose : 0, results : 0, demonstrate : 0, that : 0.53, achieves : 0.54, superior : 0, 90 : 0, compared : 0.49, traditional : 0.47, methods : 0, like : 0, k : 0, nearest : 0.49, neighbor : 0, knn : 0, however : 0, limitations : 0.47, include : 0, restricted : 0, dataset : 0.65, diversity : 0, insufficient : 0, testing : 0, under : 0, real : 0.53, world : 0, conditions : 0, recommendations : 0, emphasize : 0.48, need : 0, larger : 0.5, more : 0, diverse : 0, datasets : 0.64, integration : 0.47, additional : 0.53, media : 0, spectral : 0, data : 0.72, video : 0, enhance : 0.49, model : 0, robustness : 0 |
| 2. vvv | 130 words | 0.56 | systematic : 0, literature : 0, review : 0.5, on : 0, the : 0, application : 0, of : 0, convolutional : 0.47, neural : 0, networks : 0, for : 0, rambutan : 0, fruit : 0, classification : 0, advances : 0.49, challenges : 0, and : 0, future : 0, directions : 0, nephelium : 0, lappaceum : 0, l : 0, is : 0, a : 0, tropical : 0, widely : 0, cultivated : 0.48, in : 0, southeast : 0, asia : 0, including : 0, indonesia : 0, manual : 0, types : 0, ripeness : 0, levels : 0.5, remains : 0, challenge : 0, due : 0, to : 0, high : 0, subjectivity : 0, time : 0, intensive : 0, nature : 0, process : 0, particularly : 0, large : 0, scale : 0, agricultural : 0, operations : 0, network : 0, cnn : 0, deep : 0, learning : 0, approach : 0, offers : 0, significant : 0, potential : 0, automating : 0, improving : 0.48, accuracy : 0, tasks : 0, by : 0, extracting : 0, complex : 0, visual : 0.55, features : 0, such : 0, as : 0, color : 0, texture : 0, this : 0, study : 0, employs : 0, slr : 0, evaluate : 0.49, relevant : 0.49, research : 0, from : 0, 2019 : 0, 2024 : 0, was : 0, analyzed : 0, identify : 0, trends : 0, utilizing : 0, purpose : 0, results : 0, demonstrate : 0, that : 0, achieves : 0.49, superior : 0, 90 : 0, compared : 0, traditional : 0, methods : 0, like : 0, k : 0, nearest : 0, neighbor : 0, knn : 0, however : 0.49, limitations : 0, include : 0, restricted : 0, dataset : 0, diversity : 0.48, insufficient : 0, testing : 0, under : 0, real : 0, world : 0, conditions : 0, recommendations : 0, emphasize : 0, need : 0, larger : 0, more : 0, diverse : 0.49, datasets : 0, integration : 0, additional : 0, media : 0, spectral : 0, data : 0, video : 0.56, enhance : 0, model : 0, robustness : 0 |
Average Result
68%
0.68