堀内 隆彦

Although a physical gloss exists as a physically measurable index, people can also perceive a perceptual gloss on object surfaces. However, the physical gloss does not always match the perceptual gloss. Because the physical gloss is calculated based only on the specular gloss and does not reflect other physical features that affect the perceptual gloss. Thus, we analysed the relationships between physical features and perceptual gloss by measuring many physical properties of object surfaces, including their physical gloss. We prepared 127 flat objects comprising three materials: paper, resin and metal plating. The perceptual gloss was visually evaluated for observation angles of 20°, 60° and 85° using a magnitude estimation method. Multiple measurements were conducted to obtain physical features such as the gloss unit (GU), haze, distinctness of image and high dynamic range luminance. We then constructed prediction models for the perceptual gloss using these physical features and multiple regression analyses. By combining these multiple physical quantities and using the GU in the power scale, the prediction accuracy was improved. By the optimal power index (0.33 for physical gloss in the common prediction model, independent of the observation angle), we found that human gloss perception may be related to brightness perception.

Background and aim: Machine learning methods are examined by many researchers to identify weeds in crop images captured by drones. However, metaheuristic optimization is rarely used in optimizing the machine learning models used in weed classification. Therefore, this research targets developing a new optimization algorithm that can be used to optimize machine learning models and ensemble models to boost the classification accuracy of weed images. Methodology: This work proposes a new approach for classifying weed and wheat images captured by a sprayer drone. The proposed approach is based on a voting classifier that consists of three base models, namely, neural networks (NNs), support vector machines (SVMs), and K-nearest neighbors (KNN). This voting classifier is optimized using a new optimization algorithm composed of a hybrid of sine cosine and grey wolf optimizers. The features used in training the voting classifier are extracted based on AlexNet through transfer learning. The significant features are selected from the extracted features using a new feature selection algorithm. Results: The accuracy, precision, recall, false positive rate, and kappa coefficient were employed to assess the performance of the proposed voting classifier. In addition, a statistical analysis is performed using the one-way analysis of variance (ANOVA), and Wilcoxon signed-rank tests to measure the stability and significance of the proposed approach. On the other hand, a sensitivity analysis is performed to study the behavior of the parameters of the proposed approach in achieving the recorded results. Experimental results confirmed the effectiveness and superiority of the proposed approach when compared to the other competing optimization methods. The achieved detection accuracy using the proposed optimized voting classifier is 97.70%, F-score is 98.60%, specificity is 95.20%, and sensitivity is 98.40%. Conclusion: The proposed approach is confirmed to achieve better classification accuracy and outperforms other competing approaches.