Auto-encoder Convolut?onal Neural Network (AECNN) for Apple Fruit Flower Detection
- Title
- Auto-encoder Convolut?onal Neural Network (AECNN) for Apple Fruit Flower Detection
- Creator
- Rajagopal M.; Sathesh Kumar K.; Nagaraja P.; Sivasakthivel R.; Sivaraman G.
- Description
- The yield estimation task altogether relies upon the way toward identifying and checking the quantity of fruits on trees. In production of fruit, basic yield the board choices are guided through the bloom frequency, i.e., the quantity of the flowers that are present in a plantation. The intensity of bloom technique is still commonly assessed by methods for human visual investigation. Mechanized PC vision frameworks for flower recognizable proof depend closely on designed procedures which function just under explicit conditions and with restricted execution. This work comprises four significant advances, (I) system preparing for Fully Convolutional Network (FCN), (ii) preprocessing, (iii) component extraction, (iv) division. Initially, a strategy for assessing high-resolution pictures with deep FCN on Graphics Processing Unit (GPU). Then, non-linear and linear algorithms are presented for lessening the image noise, so the exact flower identification can be ensured. The next phase of the work handles the highlight extraction for diminishing the quality of the prime assets which are needed for handling without compromising on data applicable. By applying Local Binary Pattern (LBP), surface example likelihood can be summed up into a histogram. At last, isolate an image with high resolution into sub patches, assess all patches with the help of AECNN, at that point apply the refinement calculation on acquired score maps to figure out the final version of the mask segmentation. Trial results are led utilizing two datasets on flower pictures of AppleA and AppleB. Results are estimated regarding the measurements like Precision (P) and Recall (R). The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.
- Source
- Smart Innovation, Systems and Technologies, Vol-395 SIST, pp. 95-104.
- Date
- 2024-01-01
- Publisher
- Springer Science and Business Media Deutschland GmbH
- Subject
- Auto-encoder convolutional neural network (AECNN); Fully convolutional network (FCN); Graphics processing unit (GPU); Local binary pattern (LBP)
- Coverage
- Rajagopal M., Lean Operations and Systems, School of Business and Management, Christ (Deemed to be University), Bangalore, India; Sathesh Kumar K., Computer Science and Engineering, Alliance College of Engineering and Design, Alliance University, Central Campus, Anekal, Main Road, Karnataka, Bangalore, India; Nagaraja P., Department of Computer Science, GITAM School of Sciences, GITAM (Deemed to be University), Bangalore, India; Sivasakthivel R., Department of Computer Science, School of Sciences, Chr?st (Deemed to be University), Bangalore, India; Sivaraman G., Department of Computer Science, M.G.R. College, Tamil Nadu, Hosur, India
- Rights
- Restricted Access
- Relation
- ISSN: 21903018; ISBN: 978-981975080-1
- Format
- Online
- Language
- English
- Type
- Conference paper
Collection
Citation
Rajagopal M.; Sathesh Kumar K.; Nagaraja P.; Sivasakthivel R.; Sivaraman G., “Auto-encoder Convolut?onal Neural Network (AECNN) for Apple Fruit Flower Detection,” CHRIST (Deemed To Be University) Institutional Repository, accessed February 25, 2025, https://archives.christuniversity.in/items/show/19084.