MODEL AND HARDWARE MODULE DEVELOPMENT OF DEEP LEARNING BASED FISH APPETITE DETECTOR
Aquaculture is one of the big contributors to the food sector. The level of consumption of aquaculture products, especially fish, is projected to increase quite significantly in the next few years. The use of technology is very important to support aquaculture fish production capabilities, consideri...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/83541 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Aquaculture is one of the big contributors to the food sector. The level of consumption of aquaculture products, especially fish, is projected to increase quite significantly in the next few years. The use of technology is very important to support aquaculture fish production capabilities, considering that aquaculture efficiency is quite low as a result of operational patterns that rely solely on intuition.
This research is aimed at developing a model and hardware module for a fish appetite detection system based on multimodal deep learning which was built using input from video data and accelerometer signals. Experiments show that the addition of multimodal features can increase detection accuracy from 92% to 99%. In addition, the hardware and software results for this model also show that there is great potential for real development and implementation in the future. |
|---|