FPGA-Based Soft-Core Processors for Image Processing Applications

Infusion port level detection for intravenous infusion based on Yolo v3 neural network

PURPOSE

In order to improve the accuracy of liquid level detection in intravenous left auxiliary vein infusion and reduce the pain of patients with blood returning from intravenous infusion, we propose a deep learning based liquid level detection model of infusion levels to facilitate this operation.

METHOD

We implemented a Yolo v3-based detection model of infusion level images in intravenous infusion, and at the same time, compare it with SURF image processing technique, RCNN, and Fast-RCNN methods.

RESULTS

The model in this paper is better than the comparison algorithm in Intersection over Union (IoU), precision, recall and test time. The liquid level detection model based on Yolo v3 has a precision of 0.9768, a recall rate of 0.9688, an IoU of 0.8943, and a test time of 2.9 s.

CONCLUSION

The experimental results prove that the liquid level detection method based on deep learning has the characteristics of high accuracy and good real-time performance. This method can play a certain auxiliary role in the hospital environment and improve work efficiency of medical workers.

FPGA-Based Processor Acceleration for Image Processing Applications

FPGA-based embedded image processing systems offer considerable computing resources but present programming challenges when compared to software systems. The paper describes an approach based on an FPGA-based soft processor called Image Processing Processor (IPPro) which can operate up to 337 MHz on a high-end Xilinx FPGA family and gives details of the dataflow-based programming environment. The approach is demonstrated for a k-means clustering operation and a traffic sign recognition application, both of which have been prototyped on an Avnet Zedboard that has Xilinx Zynq-7000 system-on-chip (SoC). A number of parallel dataflow mapping options were explored giving a speed-up of 8 times for the k-means clustering using 16 IPPro cores, and a speed-up of 9.6 times for the morphology filter operation of the traffic sign recognition using 16 IPPro cores compared to their equivalent ARM-based software implementations. We show that for k-means clustering, the 16 IPPro cores implementation is 57, 28 and 1.7 times more power efficient (fps/W) than ARM Cortex-A7 CPU, nVIDIA GeForce GTX980 GPU and ARM Mali-T628 embedded GPU respectively.