Fruit freshness monitoring represents one of the key research foci in the quality control of fruits and vegetables. Traditional manual inspection methods are characterized by subjectivity and inefficiency, which renders them unsuitable for large-scale and real-time detection demands. Automated detection methods based on deep learning have increasingly attracted attention. In this study, a fruit freshness classification method based on the ResNet-101 network and a Non-local Attention mechanism is proposed. By embedding a Non-local Attention module into ResNet-101, subtle surface feature variations of the fruit are captured, thereby enhancing the model’s capacity to identify rotten areas and detect variations in color under complex backgrounds. Experimental results show that the improved model achieves a precision of 94.7%, a recall of 94.24%, and an F1-score of 94.24%, outperforming conventional ResNet-101, ResNet-50, and VGG-16 models. In particular, under complex environmental conditions, the model demonstrates significantly improved robustness in image processing. The combination of the Non-local Attention mechanism with the ResNet-101 model can substantially enhance the accuracy and stability of fruit freshness detection, which is applicable to real-time monitoring tasks in intelligent agriculture and smart logistics.
The influence of lighting color on food is a multidimensional process, linking visual interventions with people’s perception of food appearance, physiological responses, and psychological associations. This study, as a preliminary exploratory research, aims to initially investigate the effects of different lighting colors on food-induced consumer appetite and emotional perception. By measuring consumers’ physiological facial expression data, we verify whether the results are consistent with self-reported subjective evaluations. Questionnaires, Shapiro–Wilk tests, and one-sample t-tests were employed for data mining and cross-validation and combined with generalized facial expression recognition (GFER) technology to analyze participants’ emotional perceptions under various lighting colors. The results show that consumers displayed the most positive emotions and the highest appetite under 2700 K warm white light. Under this condition, the average intensity of participants’ “happy” emotion was 0.25 (SD = 0.12), indicating a clear positive emotional state. Eating willingness also reached its peak at 2700 K. In contrast, blue light-induced negative emotions and lower appetite. Among all lighting types, blue light evoked the strongest “sad” emotion (M = 0.39). This study provides a preliminary exploration of the theoretical framework regarding the relationship between food and consumer behavior, offering new perspectives for product marketing in the food industry and consumer food preference cognition. However, the generalizability of its conclusions still requires further verification in subsequent studies.
Bird feathers possess functions such as water resistance, thermal insulation, and air permeability, providing inspiration for the design of functional fabrics. Based on the functional differentiation of different feather regions and the structural constraints associated with these functions, this study selected down feathers, feather vanes, hooklets, and fluffy feather filament node structures as biomimetic prototypes. Four biomimetic knitted structures were designed for outdoor environments with significant temperature fluctuations and for the thermo-moisture comfort needs of older adults. Through macro- and micro-structural feature extraction, three-dimensional modeling, and experimental testing, a multi-parameter evaluation system covering water resistance, thermal resistance, thermal insulation rate, air permeability, moisture vapor transmission, and moisture management was established to systematically evaluate the thermo-moisture regulation performance of the fabrics. The results showed that each structure exhibited distinct performance advantages: Structure 1 demonstrated the best thermal insulation performance; Structure 2 showed relatively superior water resistance and outstanding air permeability; Structure 4 exhibited relatively superior moisture vapor transmission and moisture management performance; and Structure 3 achieved the highest gray relational optimality value, indicating a relatively balanced thermo-moisture regulation capability. Among all performance indicators, air permeability showed the highest correlation with the knitted structures. Based on these results, and considering regional differences in heat generation and sweating across different body parts of older adults, this study further explored zonal application strategies for elderly outdoor clothing to improve wearing comfort and functionality under environments with fluctuating thermal conditions.