Methods for recognizing mushroom species on the basis of the photo

Kamil Chodoła

kamil.chodola@gmail.com
Lublin University of Technology (Poland)

Grzegorz Czyż


Lublin University of Technology (Poland)

Maria Skublewska-Paszkowska


Lublin University of Technology (Poland)

Abstract

The aim of the article is to compare two methods for identifying mushroom species. In article, two methods based on one of the most popular solutions in the field of image recognition, Tenosorflow and OpenCV, have been described. A research application was created to carry out the research, in which both algorithms were implemented and tested. In addition, the application was equipped with mechanisms facilitating the collection of application data and algorithms. The results of the research have show that the method based on Tensorflow by 9% more
effectively recognizes mushroom species.


Keywords:

OpenCV;Tensorflow; image recognition

[1] Baggio D. L.: Mastering OpenCV with Practical Computer Vision Projects, Packt Publishing Ltd, 2012.
[2] Bradski G., Kaehler A.: Learning OpenCV, O’Reilly 2008.
[3] He K., Zhang X., Ren S., Sun J.: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, s. 770-778.
[4] Krzyśko M., Wołyński W., Górecki T., Skorzybut M.: Systemy uczące się, Wydawnictwo Naukowo-Techniczne Sp. z o.o., 2008.
[5] OpenCVSharp, https://github.com/shimat/OpenCVsharp [07.05.2019].
[6] Staniszewski P.: Użytkowanie grzybów leśnych – praktyka i problemy badawcze, Stud. Mater. CEPL w Rogowie, 2014, s. 143 – 152.
[7] Tang J.: Intelligent Mobile Projects with Tensorflow, Packt Publishing Ltd. 2018.
[8] TensorflowSharp, https://github.com/migueldeicaza/TensorflowSharp [11.05.2019].
Download


Published
2019-09-30

Cited by

Chodoła, K., Czyż, G., & Skublewska-Paszkowska, M. (2019). Methods for recognizing mushroom species on the basis of the photo. Journal of Computer Sciences Institute, 12, 199–205. https://doi.org/10.35784/jcsi.438

Authors

Kamil Chodoła 
kamil.chodola@gmail.com
Lublin University of Technology Poland

Authors

Grzegorz Czyż 

Lublin University of Technology Poland

Authors

Maria Skublewska-Paszkowska 

Lublin University of Technology Poland

Statistics

Abstract views: 284
PDF downloads: 288