LPPM

Permanent URI for this community

https://repository.stmik-amikbandung.ac.id/handle/0/5

Browse

Browsing LPPM by Author "Ali Suryaperdana Agoes"

Now showing 1 - 2 of 2
  • Item
    LICODS: A CNN BASED, LIGHTWEIGHT RGB-D SEMANTIC SEGMENTATION FOR OUTDOOR SCENES
    (IJICIC Editorial Office, 2019-10) Ali Suryaperdana Agoes; Zhencheng Hu; Nobutomo Matsunaga
    One way to visually understand the scenes is through per-pixel semantic segmentation. Recently, this field has undergone heavy development following the recent success of feature learning method based on the Convolutional Neural Network (CNN). Encouraged by this observation, we present Lightweight Color Depth Semantic Segmentation (LICODS), a small numbered parameter model based on the CNN for RGB-D image semantic segmentation. Additional input modality besides the color information to enhance per-pixel class prediction accuracy is employed. On the other side, our model parameter number remains low, although dual branches exist along our model’s network. The model performs better compared with the recently published RGB-D semantic segmentation models in terms of accuracy and processing time, despite of its small parameter number.
  • Item
    Object detection for KRSBI robot soccer using PeleeNet on omnidirectional camera
    (Universitas Ahmad Dahlan, 2020-08) Winarno; Ali Suryaperdana Agoes; Eva Inaiyah Agustin; Deny Arifianto
    Kontes Robot Sepak Bola Indonesia (KRSBI) is an annual event for contestants to compete their design and robot engineering in the field of robot soccer. Each contestant tries to win the match by scoring a goal toward the opponent's goal. In order to score a goal, the robot needs to find the ball, locate the goal, then kick the ball toward goal. We employed an omnidirectional vision camera as a visual sensor for a robot to perceive the object’s information. We calibrated streaming images from the camera to remove the mirror distortion. Furthermore, we deployed PeleeNet as our deep learning model for object detection. We fine-tuned PeleeNet on our dataset generated from our image collection. Our experiment result showed PeleeNet had the potential for deep learning mobile platform in KRSBI as the object detection architecture. It had a perfect combination of memory efficiency, speed and accuracy.