A METHOD FOR PROBLEM SOLVING FOGGY CITYSCAPES IMBALANCE DATASET

person suryo adhi wibowo universitas telkom
subjectAbstract
Imbalance dataset is the major problem we all will face in the process of developing deep learning model. There were many approaches to solve this very problem such as heuristic data sampling and modifying loss function for model training. In order to find the solution, we chose Foggy Cityscapes dataset for the experiment since this dataset has imbalance object class distribution. We proposed a method to solve imbalance dataset namely instance level downsampling as an extension of traditional downsampling method. The algorithm of this method will selectively keep and drop certain image in the dataset by evaluating the majority and minority object class proportion inside a single image. After comparing the model evaluation using Mean Average Precision (mAP) metric, the model which was trained with balanced dataset has more balanced knowledge or less biased across the object classes of interest.
labelKeywords:
imbalance dataset, data balancing, object detection, foggy cityscapes, deep learning
quick_reference_allReferences

T., Lin, P., Goyal, R., Girshick, K., He and P. Dollar. 2017. Focal Loss for Dense Object

Detection. Venice, Italy. IEEE International Conference on Computer Vision (ICCV).

Utomo, Syam Suryo, Cahyanto, Triawan Adi, and Prakoso, Bakhtiar Hadi. 2020. Penggunaan

Algoritma Random Over Sampling Untuk Mengatasi Masalah Imbalance Data Pada

Klasifikasi Gizi Balita. Jember, Indonesia. Universitas Muhammadiyah Jember.

Kumar, Benai. 2020. 10 Techniques to deal with Imbalanced Classes in Machine Learning.

[Online] Available at https://www.analyticsvidhya.com/blog/2020/07/10-techniques-to-dealwith-class-imbalance-in-machine-learning/ [accessed on November 28th, 2021].

Potyraj, Emily. 2021. 4 Ways to Improve Class Imbalance for Image Data. [Online] Available

at https://towardsdatascience.com/4-ways-to-improve-class-imbalance-for-image-data-

adec8f390f1 [accessed on November 29

th, 2021].

Google Developers. 2017. Imbalanced Data. [Online] Available at

https://developers.google.com/machine-learning/data-prep/construct/samplingsplitting/imbalanced-data [accessed on December 5

th, 2021].

Brems, Matt. 2020. 5 Strategies for Handling Unbalanced Classes. [Online] Available at

https://blog.roboflow.com/handling-unbalanced-classes/ [accessed on December 5

th, 2021].

Sakaridis, Christos, Dai, Dengxin, and Van Gool, Luc. 2018. Semantic Foggy Scene

Understanding with Synthetic Data. International Journal of Computer Vision. Springer

Science and Business Media LLC.

Cordts, M., Omran, M., Ramos, S., Rehfeld, et al. (2016). The Cityscapes Dataset for Semantic

Urban Scene Understanding. In Proc. of the IEEE Conference on Computer Vision and Pattern

Recognition (CVPR).

Hui, Jonathan. 2018. mAP (mean Average Precision) for Object Detection. [Online] Available

at https://jonathan-hui.medium.com/map-mean-average-precision-for-object-detection-

c121a31173 [accessed on December 22nd, 2021].

Everingham, M., Van Gool, L., Williams, C., Winn, J., and Zisserman, A. 2012. The PASCAL

Visual Object Classes Challenge 2012 (VOC2012). http://www.pascalnetwork.org/challenges/VOC/voc2012/workshop/index.html

Tan, M., Pang, R., and Le, Q.V. 2020. EfficientDet: Scalable and Efficient Object Detection.

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

Yu, Hongkun, Chen, Chen, Du, Xianzhi, Li, Yeqing, et al. (2020). TensorFlow Model Garden.

https://github.com/tensorflow/models.

licenseLicense

 

Creative Commons License
This work is licensed under a  Creative Commons Attribution-NonCommercial 4.0 International License.

Copyright Notice

An author who publishes in the Jurnal Elektro dan Telekomunikasi Terapan agrees to the following terms:

  • Author retains the copyright and grants the journal the right of first publication of the work simultaneously licensed under the Creative Commons Attribution-NonCommercial 4.0 International License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal
  • Author is able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book) with the acknowledgement of its initial publication in this journal.
  • Author is  permitted and encouraged to post his/her  work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of the published work (See The Effect of Open Access).

Read more about the Creative Commons Attribution-NonCommercial 4.0 International License. here: http://creativecommons.org/licenses/by-nc/4.0/.

Privacy Statement

The names and email addresses entered in this journal site will be used exclusively for the stated purposes of this journal and will not be made available for any other purpose or to any other party.

Most read articles by the same author(s)