Deep learning methods for drill wear classification based on images of holes drilled in melamine faced chipboard

Abstract

In this paper, a set of improvements made in drill wear recognition algorithm obtained during previous work is presented. Images of the drilled holes made on melamine faced particleboard were used as its input values. During the presented experiments, three classes were recognized: green, yellow and red, which directly correspond to a tool that is in good shape, shape that needs to be confirmed by an operator, and which should be immediately replaced, since its further use in production process can result in losses due to low product quality. During the experiments, and as a direct result of a dialog with a manufacturer it was noted that while overall accuracy is important, it is far more crucial that the used algorithm can properly distinguish red and green classes and make no (or as little as possible) misclassifications between them. The proposed algorithm is based on an ensemble of possibly diverse models, which performed best under the above conditions. The model has relatively high overall accuracy, with close to none misclassifications between indicated classes. Final classification accuracy reached 80.49% for biggest used window, while making only 7 critical errors (misclassifications between red and green classes).

Decision Confidence Assessment in Multi-Class Classification

“…The main metrics considered with the current classification model were accuracy and the number of severe (red–green) errors. In order to increase both of them, several approaches can be adapted, including using a set of subsequent images instead of single ones, as conducted in previous authors’ work [ 9 ] using window parameters. After that, the dominant value of all classifications can be used.…”

“…There are also some recent approaches that incorporate similar methodologies. In [ 9 ], various CNN networks were tested and evaluated to prepare an improved approach that focused more on limiting critical errors that the classifier makes. In another solution, presented in [ 10 ], a Siamese network was applied to the same problem, which is a new, CNN-based methodology.…”

The Review of New Scientific Developments in Drilling in Wood-Based Panels with Particular Emphasis on the Latest Research Trends in Drill Condition Monitoring

“…Generally, tool condition monitoring in the field of woodworking has also been popular for a long time [24][25][26]. Therefore, at the end of this introductory (and as concisely as possible) overview of the latest research trends, it is also worth noting the new and quite spectacular approach to drill condition monitoring in wood-based panels machining [27][28][29][30][31][32][33][34][35][36][37][38].…”

“…In the last 2 or 3 years, numerous research and scientific analyses have been carried out to address this problem. Some of them seem to be truly innovative and their synthetic review is the main content of this paper [33][34][35][36][37][38]. The purpose is to encourage more scientists to cooperate or compete in this research area (both attitudes are beneficial from a scientific development point of view) currently.…”

Multiclass Image Classification Using GANs and CNN Based on Holes Drilled in Laminated Chipboard

“…In the latest work on drill wear recognition [ 15 ], a set of models were compared to classify images into three classes. It was proved that due to unsatisfactory results for three classes of drill wear recognition, a division into two classes is sufficient from the business perspective, as misclassifying the worst class with the best class generates the greatest loss.…”