Quantifying plastic refuse in water area helps to understand how plastic refuse accumulates in water area and is essential for targeted cleanup efforts. Currently, the most common methods for quantifying plastic in water area are human visual counting and sampling using nets, but such methods are costly and labor-intensive. This study proposes a watershed refuse identification algorithm based on an improved YOLOv4. Lightweight improvements to YOLOv4. EfficientNetB1 is used to replace the backbone network of YOLOv4, and the Depthwise Convolution is used to replace the original convolution to reduce the number of model parameters and computation. The anchors are re-clustered using k-means algorithm to improve the accuracy. The experimental results show that the improved algorithm improves the detection speed by 11.2% and reduces the number of parameters by 76.54% compared with YOLOv4 at the expense of 0.69% recognition accuracy.

This system uses temperature sensor DS18B20, Bluetooth module HC-05, microcontroller STM32f103ZET6, LCD display with 8080 port, RTC clock and alarm circuit to achieve portable temperature measurement rapidly. It can accurately measure the temperature and the deviation is kept within 0.5 degrees. The collected temperature data and the current time are displayed on the display screen. Send the collected temperature data at regular time and receive by APP mobile phone. When users want to get temperature data remotely on their phones, press the send button on the touch screen, press the up button and send the “wendu#” command to the cell phone APP to get the temperature data collected. When the temperature is over 37∘C, the four LED lights will always be on and LCD will show “Temperature is too high, please pay attention!” This design can not only realize the function of temperature measurement simply, but also optimize the display interface, modify the system time on the display interface to achieve the humanized function, which is convenient for daily portable use.