Smart Color Assistant for the Visually Impaired Based on TCS34725 Sensor and KNN Algorithm
DOI:
https://doi.org/10.22303/csrid-.18.1.2026.1-17Keywords:
Visually Impaired, TCS34725 Color Sensor, KNN (K-Nearest Neighbor), ESP32, Smart Color Assistant, Portable Assistive DeviceAbstract
The ability to recognize colors is an important aspect of human interaction with the environment, but it poses a significant challenge for people with visual impairments. This study aims to design and implement a prototype portable device called Smart Color Assistant to help people with visual impairments recognize colors. The system was developed based on an ESP32 microcontroller integrated with a TCS34725 color sensor for RGB data acquisition and a DFPlayer Mini module for sound output. The KNN (K-Nearest Neighbor) algorithm with a K value of 5 was applied to classify 16 common colors. The research approach applied was the Research and Development (R&D) method, which included several stages, namely needs analysis, system design, hardware and software implementation, and testing and evaluation processes. Based on the test results, it was found that the TCS34725 sensor had an average reading error of 2.61%. The system achieved a classification accuracy of 93.75% (15 out of 16 colors were detected correctly), with an average response time from detection to sound output of 850 milliseconds. Functionality testing of the buttons and sound output also showed performance in line with the design. Testing with visually impaired individuals confirmed the benefits of the device, although improvements are needed in terms of ergonomics and ease of use. It is concluded that this prototype has the potential to be an effective, portable, and responsive assistive device.
References
B. A. Ashad, R. Ramdaniah, S. Sriwijanaka, dan St. H. Mansyur, “Pemanfaatan Aplikasi Pendeteksi Warna Pakaian Berbasis Android bagi Penyandang Tunanetra di SLB Yukartuni Makassar,” J. Abdi Masy. Indones., vol. 2, no. 1, hlm. 263–268, Jan 2022, doi: 10.54082/jamsi.191.
R. F. Abadi dkk., “Efektivitas penggunaaan alat pendeteksi warna terhadap kemampuan bina diri tunanetra,” J. UNIK Pendidik. Luar Biasa, vol. 7, no. 1, hlm. 19, Jul 2022, doi: 10.30870/unik.v7i1.6911.
A. G. Ginting dan Y. Siyamto, “ALAT PENDETEKSI PENGECEKAN KEMATANGAN BUAH TOMAT MENGGUNAKAN ARDUINO DENGAN SENSOR WARNA,” no. 5, 2021.
N. Othman, M. Z. Md. Zain, I. Shah Ishak, A. R. Abu Bakar, M. Ab Wahid, dan M. Mohamad, “A colour recognition device for the visually disabled people,” Indones. J. Electr. Eng. Comput. Sci., vol. 17, no. 3, hlm. 1322, Mar 2020, doi: 10.11591/ijeecs.v17.i3.pp1322-1329.
M. A. Wahid, M. Y. Ahmad, M. Z. Zain, S. M. Saad, M. Hussein, dan Z. Ahmad, “A Preliminary Study on Colour Recogniser for Visually Disabled People”.
F. Anwar, “Application of Digital Cellular Radio for Mobile Location Estimation,” IIUM Eng. J., vol. 2, no. 2, Jan 1970, doi: 10.31436/iiumej.v2i2.345.
Hilman Aziz dan Imam Suharjo, “Pengembangan Sistem Keamanan Gerbang Rumah Smart Home Berbasis IoT dengan Metode RnD,” JEKIN - J. Tek. Inform., vol. 4, no. 3, hlm. 663–674, Agu 2024, doi: 10.58794/jekin.v4i3.895.
A. E. Kumara dan J. Hendry, “Purwarupa Light Meter Menggunakan Sensor Warna untuk Fotografi,” J. Pekommas, vol. 9, no. 1, hlm. 121–130, Jun 2024, doi: 10.56873/jpkm.v9i1.5569.
R. N. Fitasari dan D. Irawan, “Prototype of Tomato Fruit Sorting Based on Color Based on Esp32-Cam with Fuzzy Logic Method,” vol. 12, no. 1, 2025.
M. Suyal dan P. Goyal, “A Review on Analysis of K-Nearest Neighbor Classification Machine Learning Algorithms based on Supervised Learning,” Int. J. Eng. Trends Technol., vol. 70, no. 7, hlm. 43–48, Jul 2022, doi: 10.14445/22315381/IJETT-V70I7P205.
M. W. Lestari, “DESIGN AND CONSTRUCTION OF A NATURAL DISASTER DETECTION EQUIPMENT USING THE WATER LEVEL SWITCH SENSOR AND SENSOR SW-420,” vol. 1, no. 2.
I. G. Romagusta, E. Sonalitha, dan S. Subairi, “Prediksi Kematangan Tomat Menggunakan Sensor Warna dengan Metode KNN (K-Nearest Neighbor),” Blend Sains J. Tek., vol. 3, no. 1, hlm. 92–101, Agu 2024, doi: 10.56211/blendsains.v3i1.600.
F. T. Admojo dan Ahsanawati, “Klasifikasi Aroma Alkohol Menggunakan Metode KNN,” Indones. J. Data Sci., vol. 1, no. 2, hlm. 34–38, Jul 2020, doi: 10.33096/ijodas.v1i2.12.
T. P. Putera, A. Angraini, E. Saputra, M. Megawati, M. Fronita, dan A. Marsal, “Usability Evaluation of Pekanbaru dalam Genggaman Application using System Usability Scale (SUS),” SISTEMASI, vol. 13, no. 2, hlm. 608, Mar 2024, doi: 10.32520/stmsi.v13i2.3782.
E. Roshan dan B. Funt, “Color Sensor Accuracy Index Utilizing Metamer Mismatch Radii,” Sensors, vol. 20, no. 15, hlm. 4275, Jul 2020, doi: 10.3390/s20154275.
T.-K. Woodstock dan R. F. Karlicek, “RGB Color Sensors for Occupant Detection: An Alternative to PIR Sensors,” IEEE Sens. J., vol. 20, no. 20, hlm. 12364–12373, Okt 2020, doi: 10.1109/JSEN.2020.3000170.
L. Farokhah, “Implementasi K-Nearest Neighbor untuk Klasifikasi Bunga Dengan Ekstraksi Fitur Warna RGB,” J. Teknol. Inf. Dan Ilmu Komput., vol. 7, no. 6, hlm. 1129–1136, Des 2020, doi: 10.25126/jtiik.2020722608.
