Comparative Evaluation of Random Forest and CNN for Shape Classification in a Simulated Robotic Sorting Application

Author

Bogdan Sofian, George-Alin Andrei, Rafael Patrick Patruti


Abstract

 This paper compares Random Forest and a Convolutional Neural Network (CNN) for shape classification in a simulated robotic sorting application. A synthetic dataset containing 1,100 images was generated with cubes, cylinders, and cones in three colors. The models were evaluated using 150 test images containing 377 objects. Random Forest achieved 99.73% accuracy on individual object patches, while CNN achieved 100%. In the complete image processing pipeline, the shape classification accuracies for matched detections were 98.01% and 99.72%, respectively. Both methods achieved a detection F1-score of 95.39% because they used the same object detection procedure. A robotic sorting prototype was also implemented in CoppeliaSim using a KUKA LBR iiwa manipulator and a BarrettHand gripper. The results show that CNN provided better shape recognition on the synthetic dataset. Future work will focus on images captured directly from the simulator and validation of complete robotic sorting operations.



Keywords

CoppeliaSim; Python; Robotic Sorting; Computer Vision; Convolutional Neural Network; Random Forest.



Full Text:

Download Paper PDF


References


  1. Santos, A. A., Schreurs, C., da Silva, A. F., Pereira, F., Felgueiras, C., Lopes, A. M., & Machado, J. (2024). Integration of artificial vision and image processing into a pick and place collaborative robotic system. Journal of Intelligent & Robotic Systems, 110, 159. https://doi.org/10.1007/s10846-024-02195-z
  2. Drăgoi, M.-V., Cristoiu, C. A., Nechita, R.-M., Navligu, B.-C., & Verdete, B.-M. (2026). A Python-based framework for learning-from-demonstration in robotic object sorting: Comparative evaluation of lightweight classifiers. Applied Sciences, 16(10), 5107. https://doi.org/10.3390/app16105107
  3. Hong, F., Tay, D. W. L., & Ang, A. (2023). Intelligent pick-and-place system using MobileNet. Electronics, 12(3), 621. https://doi.org/10.3390/electronics12030621
  4. Cristoiu, C. A., Drăgoi, M.-V., Cojocaru, A., & Spânu, P. (2026). A vision-based software safety monitoring tool for operators in RoboDK robotic cells: A simulation-based proof-of-concept study using workspace masks and image processing. Technologies, 14(6), 373. https://doi.org/10.3390/technologies14060373
  5. Tălîngă, A.-M., Hadăr, A., Drăgoi, M.-V., Nisipeanu, I., Ali, H. A., & Suciu, C. P. (2024). YOLO-v8 in capturing imperfections generated by changing 3D printer parameters. U.P.B. Scientific Bulletin, Series C, 86(4), 253–266. https://www.scientificbulletin.upb.ro/rev_docs_arhiva/full72b_799902.pdf
  6. Schieber, H., Demir, K. C., Kleinbeck, C., Yang, S. H., & Roth, D. (2024). Indoor synthetic data generation: A systematic review. Computer Vision and Image Understanding, 240, 103907. https://doi.org/10.1016/j.cviu.2023.103907
  7. Drăgoi, M.-V., & Cristoiu, C. A. (2026). A review of the fused deposition modeling (FDM) process: An integrated analysis of materials, manufacturing parameters, mechanical properties, defect formation, and quality assessment. International Journal of Research Publication and Reviews, 7(10), 4807–4817. https://doi.org/10.55248/gengpi.07.1026.33106
  8. Tache, F.-G., Drăgoi, M.-V., Puiu, R.-A., Petrea, G., & Pascaru, R.-C. (2026). Application in research projects on the innovation found in the field of engineering of polymer and composites products. Macromolecular Symposia, 415(3), e70379. https://doi.org/10.1002/masy.70379
  9. Borrell, J., Perez-Vidal, C., & Segura, J. V. (2024). Optimization of the pick-and-place sequence of a bimanual collaborative robot in an industrial production line. The International Journal of Advanced Manufacturing Technology, 130, 4221–4234. https://doi.org/10.1007/s00170-023-12922-9
  10. Cristoiu, C. A., Drăgoi, M.-V., Nechita, R.-M., Verdete, B.-M., Dudici, C. L., & Cusma, C. N. (2026). Python software application for obstacle-avoiding path planning in RoboDK using free space graph and robot level validation. Applied Sciences, 16(10), 4786. https://doi.org/10.3390/app16104786
  11. Cristoiu, C.-A., Drăgoi, M.-V., & Georgescu, V.-C. (2026). Comparative analysis of Cartesian, cylindrical and spherical grids in a graph-based obstacle-avoidance planner for industrial robots. Applied Sciences, 16(12), 6189. https://doi.org/10.3390/app16126189
  12. Cristoiu, C. A., Drăgoi, M.-V., Nechita, R.-M., Stoian, B. N., Dudici, C. L., Puiu, R.-A., & Petrea, G. (2026). The impact of geometric continuities (C1, C2, and C3) on the trajectories of industrial robots. Technologies, 14(3), 176. https://doi.org/10.3390/technologies14030176
  13. Cristoiu, C. A., Drăgoi, M.-V., Ivan, A. M., Nechita, R.-M., Grecu, I., Puiu, R.-A., Petrea, G., & Popescu, E. (2025). Optimization method based on the minimum action principle for trajectory length of articulated manipulators. Technologies, 13(11), 490. https://doi.org/10.3390/technologies13110490
  14. Cristoiu, C., Ivan, M., Dragoi, M. V., Popescu, E., & Nechita, R.-M. (2025). Method for qualitative evaluation of robotic trajectories based on geometric and kinematic metrics. Proceedings in Manufacturing Systems, 20(2), 69–74.
  15. Drăgoi, M.-V., Frimu, A.-V., Postelnicu, A., Puiu, R.-A., Petrea, G., & Hank, A. (2026). Interactive teleoperation of an articulated robotic arm using vision-based human hand tracking. Biomimetics, 11(2), 151. https://doi.org/10.3390/biomimetics11020151
  16. Jain, J., Iyer, S. V., Sangwan, K. S., & Dhiraj. (2025). Vision-driven robotic grasping order generation using segmentation and relative positioning in a cluttered environment. Procedia Computer Science, 258, 2563–2572. https://doi.org/10.1016/j.procs.2025.04.518
  17. Moyano, J., Musicco, A., Nieto-Julián, J. E., & Domínguez-Morales, J. P. (2024). Geometric characterization and segmentation of historic buildings using classification algorithms and convolutional networks in HBIM. Automation in Construction, 167, 105728. https://doi.org/10.1016/j.autcon.2024.105728
  18. Drăgoi, M.-V., Afram, R. M., Alzubaidi, A. A. A. M. H., Jiga, G. G., Ali, H. A., Puiu, R.-A., Petrea, G., & Hank, A. (2026). Comparative analysis of interpolation and extrapolation methods in regression models. U.P.B. Scientific Bulletin, Series C, 88(1), 145–160. https://www.scientificbulletin.upb.ro/seria/c/arhiva?year=2026
  19. Loghin, A., Badea, F., & Tamaș, E. A. (2026). A reliability-aware Edge AI application for industrial machine failure prediction under degraded sensor inputs. International Journal of Mechatronics & Applied Mechanics, 1(25), 145–157. https://doi.org/10.17683/ijomam/issue25/v1.12
  20. Buzducea, C.-A., Drăgoi, M.-V., Cristoiu, C., Puiu, R.-A., Puiu, M., Petrea, G., & Navligu, B.-C. (2026). Machine learning in education: Predicting student performance and guiding institutional decisions. Education Sciences, 16(1), 76. https://doi.org/10.3390/educsci16010076
  21. Puiu, R.-A., Drăgoi, M.-V., Tache, F.-G., Spiridon-Mocioacă, T.-M., Petrea, G., & Hank, A. (2026). Interactive attrition prediction for academic human resources management using machine learning. U.P.B. Scientific Bulletin, Series C, 88(2), 217–232. https://www.scientificbulletin.upb.ro/static/pdfs/a2026nr2c_16_16653_Roxana_Puiu_2_pag.pdf
  22. Drăgoi, M.-V., Puiu, R.-A., Petrea, G., Cristoiu, C. A., & Dumitrescu, C.-I. (2026). Willingness to allow educational data use for learning analytics in higher education: Trust and governance predictors: An exploratory study. Education Sciences, 16(4), 637. https://doi.org/10.3390/educsci16040637
  23. Drăgoi, M., Buzducea, C., Marin, A., & Boanță, L. (2026). Assessment pressure mapper for visualizing peaks in digital evaluation load. FAIMA Business & Management Journal, 14(1), 34–42. https://faimajournal.upb.ro/archive/RevistaFAIMA_Martie2026.pdf
  24. Drăgoi, M.-V., Puiu, R.-A., Petrea, G., Burtea, C., Spiridon-Mocioacă, T.-M., Puiu, M., & Suciu, C. P. (2025). Improving the learning process for students. Personalizing learning through software applications. U.P.B. Scientific Bulletin, Series C, 87(1), 157–166. https://www.scientificbulletin.upb.ro/rev_docs_arhiva/full465_531870.pdf
  25. Drăgoi, M.-V., Nisipeanu, I., Puiu, R.-A., Tache, F.-G., Spiridon-Mocioacă, T.-M., Hank, A., & Cristoiu, C. (2025). An inclusive offline learning platform integrating gesture recognition and local AI models. Biomimetics, 10(10), 693. https://doi.org/10.3390/biomimetics10100693
  26. Okafor, E., Oyedeji, M., & Alfarraj, M. (2024). Deep reinforcement learning with light-weight vision model for sequential robotic object sorting. Journal of King Saud University - Computer and Information Sciences, 36(1), 101896. https://doi.org/10.1016/j.jksuci.2023.101896
  27. Drăgoi, M.-V., Nisipeanu, I., Marin, I., Cristoiu, C. A., & Alexe, C.-G. (2026). Bio-inspired gaze and neural command fusion for assistive smartphone interaction. Biomimetics, 11(7), 514. https://doi.org/10.3390/biomimetics11070514
  28. Drăgoi, M.-V., Nisipeanu, I., Frimu, A., Tălîngă, A.-M., Hadăr, A., Dobrescu, T. G., Suciu, C. P., & Manea, A. R. (2024). Real-time home automation system using BCI technology. Biomimetics, 9(10), 594. https://doi.org/10.3390/biomimetics9100594
  29. Ali, H. A., Goga, N., Vasilateanu, A., Muslim, A. M., Ali, K. A., & Drăgoi, M.-V. (2022). English and Romanian brain-to-text brain-computer interface word prediction system. International Journal of Advanced Computer Science and Applications, 13(10). https://doi.org/10.14569/IJACSA.2022.0131003
  30. Drăgoi, M.-V., Hadăr, A., Goga, N., Grigore, L. Ș., Ștefan, A., & Ali, H. A. (2023). Design and implementation of an EEG-based BCI prosthetic lower limb using Raspberry Pi 4. U.P.B. Scientific Bulletin, Series C, 85(3), 353–366. https://www.scientificbulletin.upb.ro/static/pdfs/full704_649223.pdf
  31. Drăgoi, M.-V., Hadăr, A., Goga, N., & Ali, H. A. (2022). Determining the user's requirements for an intelligent mind-controlled prosthetic lower limb system. In 2022 IEEE International Conference on Blockchain, Smart Healthcare and Emerging Technologies (SmartBlock4Health) (pp. 1–6). IEEE. https://doi.org/10.1109/SmartBlock4Health56071.2022.10034519
  32. Drăgoi, M.-V., Hadăr, A., Goga, N., Baciu, F., Ștefan, A., Grigore, L. Ș., Gorgoteanu, D., Molder, C., & Oncioiu, I. (2023). Contributions to the dynamic regime behavior of a bionic leg prosthesis. Biomimetics, 8(5), 414. https://doi.org/10.3390/biomimetics8050414
  33. Goga, N., Boiangiu, C.-A., Vasilateanu, A., Popovici, A.-F., Drăgoi, M.-V., Popovici, R., Gancea, I. O., Pîrlog, M. C., Popa, R. C., & Hadăr, A. (2022). An efficient system for eye movement desensitization and reprocessing (EMDR) therapy: A pilot study. Healthcare, 10(1), 133. https://doi.org/10.3390/healthcare10010133
  34. Gancea, I. O., Drăgoi, M.-V., Popovici, A.-F., Goga, N., Dragomir, R., Goga, M., & Dinu, M. C. (2020). An intelligent EMDR system for helping students with psychological problems. eLearning & Software for Education, 2, 207–214. https://doi.org/10.12753/2066-026X-20-111
  35. Chen, C., Wang, J., Liu, J., Du, G., Zhu, X., Guan, Q., & Qiu, X. (2026). A framework for vision-based robotic grasping and sorting in cluttered environments. Industrial Robot, 53(1), 102–113. https://doi.org/10.1108/IR-12-2024-0533
  36. Drăgoi, M.-V., Cristoiu, C. A., & Tamaș, E. A. (2026). Industry 5.0 and sustainable business models in digital transformation. FAIMA Business & Management Journal, 14(3), 57–67. https://faimajournal.upb.ro/archive/RevistaFaima_11sept2026_final.pdf
  37. Drăgoi, M.-V., Șerban, D. O., Cristoiu, C. A., Boanță, L. F., & Marin, A. (2026). Hybrid project management readiness in Romanian construction companies: An exploratory survey from a small construction-company context. Buildings, 16(19), 3877. https://doi.org/10.3390/buildings16193877
  38. Tache, F.-G., Drăgoi, M.-V., Puiu, R.-A., Petrea, G., & Pascaru, R.-C. (2025). A decision-support tool for evaluating employee contribution toward organizational excellence. In International Conference of Management and Industrial Engineering (Vol. 12, pp. 345-352). https://doi.org/10.56177/12icmie2025.109
  39. Buzducea, C.-A., Tache, F.-G., Puiu, M., & Semenescu, A. (2026). Adaptation of innovative tools for human resource management in international research projects. Proceedings of the International Conference on Business Excellence, 20(1), 1507–1516. https://doi.org/10.2478/picbe-2026-0115
  40. Nechita, R.-M., Deselnicu, D.-C., Băjenaru, V.-D., Istrițeanu, S.-E., Cristoiu, C., & Drăgoi, M.-V. (2026). Optimizing material recovery from photovoltaic waste: A performance ranking using hybrid BWM-PROMETHEE II. Sustainability, 18(8), 3750. https://doi.org/10.3390/su18083750
  41. Tălîngă, A.-M., Baciu, F., Hadăr, A., Jiga, G., Drăgoi, M.-V., & Ancuța, A. (2025). Numerical and experimental study of a sandwich beam subjected to a three points bending test. U.P.B. Scientific Bulletin, Series D, 87(3), 227–240. https://www.scientificbulletin.upb.ro/static/pdfs/fulld2b_804128.pdf
  42. Drăgoi, M.-V., Maier, A. N., Alzubaidi, A. A. A. M. H., Suciu, C. P., Ali, H. A., Dobrescu, T. G., Hadăr, A., Puiu, R.-A., Petrea, G., & Avăsiloaie, A. P. (2025). Multithreading VPN application to monitor web traffic and unblock geoblocking websites. U.P.B. Scientific Bulletin, Series C, 87(4). https://www.scientificbulletin.upb.ro/static/pdfs/rez17b_486740.pdf
  43. Bolovan, M.-A., Lupșa, Ș.-A., & Baltag, C. A. (2026). Python-Based Color and Shape Sorting in Coppelia Sim: An OpenCV and Random Forest Case Study. International Journal of Research Publication and Reviews, 7(10), 9409–9415. https://doi.org/10.55248/gengpi.07.1026.35014

 

 

Share your valuable work from Social Media Buttons