Wearable Bioimpedance-Based Deep Learning Techniques for Live Fish Health Assessment under Waterless and Low-Temperature Conditions.

(1) Background: At present, physiological stress detection technology is a critical means for precisely evaluating the comprehensive health status of live fish. However, the commonly used biochemical tests are invasive and time-consuming and cannot simultaneously monitor and dynamically evaluate multiple stress levels in fish and accurately classify their health levels. The purpose of this study is to deploy wearable bioelectrical impedance analysis (WBIA) sensors on fish skin to construct a deep learning-based stress dynamic evaluation model for precisely estimating their accurate health status. (2) Methods: The correlation of fish (turbot) muscle nutrients and their stress indicators are calculated using grey relation analysis (GRA) for allocating the weight of the stress factors. Next, WBIA features are sieved using the maximum information coefficient (MIC) in stress trend evaluation modeling, which is closely related to the key stress factors. Afterward, a convolutional neural network (CNN) is utilized to obtain the features of the WBIA signals. Then, the long short-term memory (LSTM) method learns the stress trends with residual rectification using bidirectional gated recurrent units (BiGRUs). Furthermore, the Z-shaped fuzzy function can accurately classify the fish health status by the total evaluated stress values. (3) Results: The proposed CNN-LSTM-BiGRU-based stress evaluation model shows superior accuracy compared to the other machine learning models (CNN-LSTM, CNN-GRU, LSTM, GRU, SVR, and BP) based on the MAPE, MAE, and RMSE. Moreover, the fish health classification under waterless and low-temperature conditions is thoroughly verified. High accuracy is proven by the classification validation criterion (accuracy, F1 score, precision, and recall). (4) Conclusions: the proposed health evaluation technology can precisely monitor and track the health status of live fish and provides an effective technical reference for the field of live fish vital sign detection.

Millimeter Wave-Based Non-Destructive Biosensor System for Live Fish Monitoring.

Waterless transportation for live grouper is a novel mode of transport that not only saves money, but also lowers wastewater pollution. Technical obstacles remain, however, in achieving intelligent monitoring and a greater survival rate. During live grouper waterless transportation, the stress response is a key indicator that affects the survival life-span of the grouper. Studies based on breathing rate analysis have demonstrated that among many stress response parameters, breathing rate is the most direct parameter to reflect the intensity. Conventional measurement methods, which set up sensors on the gills of groupers, interfere with the normal breathing of living aquatic products and are complex in system design. We designed a new breathing monitoring system based on a completely non-destructive approach. The system allows the real-time monitoring of living aquatic products' breathing rate by simply placing the millimeter wave radar on the inner wall of the incubator and facing the gills. The system we developed can detect more parameters in the future, and can replace the existing system to simplify the study of stress responses.

Deep-Learning-Based Wireless Visual Sensor System for Shiitake Mushroom Sorting.

The shiitake mushroom is the second-largest edible mushroom in the world, with a high nutritional and medicinal value. The surface texture of shiitake mushrooms can be quite different due to different growing environments, consequently leading to fluctuating market prices. To maximize the economic profit of the mushroom industry, it is necessary to sort the harvested mushrooms according to their qualities. This paper aimed to develop a deep-learning-based wireless visual sensor system for shiitake mushroom sorting, in which the visual detection was realized by the collection of images and cooperative transmission with the help of visual sensors and Wi-Fi modules, respectively. The model training process was achieved using Vision Transformer, then three data-augmentation methods, which were Random Erasing, RandAugment, and Label Smoothing, were applied under the premise of a small sample dataset. The training result of the final model turned out nearly perfect, with an accuracy rate reaching 99.2%. Meanwhile, the actual mushroom-sorting work using the developed system obtained an accuracy of 98.53%, with an 8.7 ms processing time for every single image. The results showed that the system could efficiently complete the sorting of shiitake mushrooms with a stable and high accuracy. In addition, the system could be extended for other sorting tasks based on visual features. It is also possible to combine binocular vision and multisensor technology with the current system to deal with sorting work that requires a higher accuracy and minor feature identification.

Development and evaluation of an intelligent traceability system for frozen tilapia fillet processing.

BACKGROUND: The main export varieties in China are brand-name, high-quality bred aquatic products. Among them, tilapia has become the most important and fast-growing species since extensive consumer markets in North America and Europe have evolved as a result of commodity prices, year-round availability and quality of fresh and frozen products. As the largest tilapia farming country, China has over one-third of its tilapia production devoted to further processing and meeting foreign market demand. RESULTS: Using by tilapia fillet processing, this paper introduces the efforts for developing and evaluating ITS-TF: an intelligent traceability system integrated with statistical process control (SPC) and fault tree analysis (FTA). Observations, literature review and expert questionnaires were used for system requirement and knowledge acquisition; scenario simulation was applied to evaluate and validate ITS-TF performance. CONCLUSION: The results show that traceability requirement is evolved from a firefighting model to a proactive model for enhancing process management capacity for food safety; ITS-TF transforms itself as an intelligent system to provide functions on early warnings and process management by integrated SPC and FTA. The valuable suggestion that automatic data acquisition and communication technology should be integrated into ITS-TF was achieved for further system optimization, perfection and performance improvement.

MS-BWME: a wireless real-time monitoring system for brine well mining equipment.

This paper describes a wireless real-time monitoring system (MS-BWME) to monitor the running state of pumps equipment in brine well mining and prevent potential failures that may produce unexpected interruptions with severe consequences. MS-BWME consists of two units: the ZigBee Wireless Sensors Network (WSN) unit and the real-time remote monitoring unit. MS-BWME was implemented and tested in sampled brine wells mining in Qinghai Province and four kinds of indicators were selected to evaluate the performance of the MS-BWME, i.e., sensor calibration, the system's real-time data reception, Received Signal Strength Indicator (RSSI) and sensor node lifetime. The results show that MS-BWME can accurately judge the running state of the pump equipment by acquiring and transmitting the real-time voltage and electric current data of the equipment from the spot and provide real-time decision support aid to help workers overhaul the equipment in a timely manner and resolve failures that might produce unexpected production down-time. The MS-BWME can also be extended to a wide range of equipment monitoring applications.