Research and implementation of a fabric printing detection system based on a field programmable gate array and deep neural network

2021 ◽  
pp. 004051752110481
Author(s):  
Feng Li ◽  
Qinggang Xi
Keyword(s):  
Neural Network ◽  
Field Programmable Gate Array ◽  
Detection System ◽  
Parallel Optimization ◽  
Detection Accuracy ◽  
Processing Unit ◽  
Backbone Network ◽  
The Neural Network ◽  
Field Programmable ◽  
Gate Array

In this paper, aiming at the problems of difficult positioning, slow speed and low precision of digital printing, a detection system suitable for textile printing positioning is proposed and designed. This detection system innovatively combines a neural network and field programmable gate array (FPGA) to realize rapid and accurate positioning of printing. In the neural network part, this paper selects the backbone network Darknet19 of YOLOv2 as the backbone network, and under the premise of ensuring a certain detection accuracy, the network model is pruned and quantified to make it suitable for deployment on the embedded device FPGA. In addition, before the network training, this paper optimizes the candidate boxes by introducing k-means clustering to customize the analysis of the fabric print dataset to improve the detection accuracy. In the FPGA part, this paper optimizes the architecture on the FPGA side in two parts: data computation and data transmission. In terms of computational optimization, parallel optimization of the neural network is performed by combining FPGA optimization methods, such as pipeline and unroll. In terms of transmission optimization, we use a double-buffered design to ping-pong in the input and output modules to overlap the latency, and then use multi-port transmission to improve the overall bandwidth utilization and reduce the transmission latency caused by on-chip and off-chip interactions. The experimental results show that the detection system combining the neural network and FPGA can effectively position fabric prints and meet the needs of real-time. The design scheme has lower power compared to the graphics processing unit and is faster compared to the central processing unit.

scholarly journals Real-time object tracking system based on field-programmable gate array and convolution neural network

2016 ◽  
Vol 14 (1) ◽  
pp. 172988141668270 ◽  
Author(s):  
Congyi Lyu ◽  
Haoyao Chen ◽  
Xin Jiang ◽  
Peng Li ◽  
Yunhui Liu
Keyword(s):  
Neural Network ◽  
Image Processing ◽  
Object Tracking ◽  
Real Time ◽  
Field Programmable Gate Array ◽  
Tracking System ◽  
Convolution Neural Network ◽  
Processing Unit ◽  
Field Programmable ◽  
Gate Array

Vision-based object tracking has lots of applications in robotics, like surveillance, navigation, motion capturing, and so on. However, the existing object tracking systems still suffer from the challenging problem of high computation consumption in the image processing algorithms. The problem can prevent current systems from being used in many robotic applications which have limitations of payload and power, for example, micro air vehicles. In these applications, the central processing unit- or graphics processing unit-based computers are not good choices due to the high weight and power consumption. To address the problem, this article proposed a real-time object tracking system based on field-programmable gate array, convolution neural network, and visual servo technology. The time-consuming image processing algorithms, such as distortion correction, color space convertor, and Sobel edge, Harris corner features detector, and convolution neural network were redesigned using the programmable gates in field-programmable gate array. Based on the field-programmable gate array-based image processing, an image-based visual servo controller was designed to drive a two degree of freedom manipulator to track the target in real time. Finally, experiments on the proposed system were performed to illustrate the effectiveness of the real-time object tracking system.

scholarly journals Implementación de redes neuronales en FPGAs usando tipos de datos de punto fijo

2020 ◽  
Vol 8 ◽  
Author(s):  
Daniel Enériz Orta ◽  
Nicolás Medrano Marqués ◽  
Belén Calvo López
Keyword(s):  
Field Programmable Gate Array ◽  
A Priori ◽  
Central Processing Unit ◽  
Processing Unit ◽  
Central Processing ◽  
Field Programmable ◽  
Gate Array

La capacidad de estimar funciones no lineales hace que las redes neuronales sean una de las herramientas más usadas para aplicar fusión sensorial, permitiendo combinar la salida de diferentes sensores para obtener información de la que a priori no se dispone. Por otra parte, la capacidad de procesamiento paralelo de las FPGAs (Field-Programmable Gate Array) las hace idóneas para implementar redes neuronales ubicuas, permitiendo inferir resultados más rápido que una CPU (Central Processing Unit) sin necesidad de una conexión activa a internet. De esta forma, en este artículo se propone un flujo de trabajo para diseñar, entrenar e implementar una red neuronal en una FPGA Xilinx PYNQ Z2 que use tipos de dato de punto fijo para hacer fusión sensorial. Dicho flujo de trabajo es probado mediante el desarrollo de una red neuronal que combine las salidas de una nariz artificial de 16 sensores para obtener una estimación de las concentraciones de CH4 y C2H4.

A field-programmable gate array system for sonar image recognition based on convolutional neural network

2020 ◽  
pp. 095965182093934
Author(s):  
Chong Wang ◽  
Yu Jiang ◽  
Kai Wang ◽  
Fenglin Wei
Keyword(s):  
Neural Network ◽  
Energy Consumption ◽  
Convolutional Neural Network ◽  
Image Recognition ◽  
Field Programmable Gate Array ◽  
Oil And Gas ◽  
Sonar Image ◽  
Field Programmable ◽  
Sonar Images ◽  
Gate Array

Subsea pipeline is the safest, most reliable, and most economical way to transport oil and gas from an offshore platform to an onshore terminal. However, the pipelines may rupture under the harsh working environment, causing oil and gas leakage. This calls for a proper device and method to detect the state of subsea pipelines in a timely and precise manner. The autonomous underwater vehicle carrying side-scan sonar offers a desirable way for target detection in the complex environment under the sea. As a result, this article combines the field-programmable gate array, featuring high throughput, low energy consumption and a high degree of parallelism, and the convolutional neural network into a sonar image recognition system. First, a training set was constructed by screening and splitting the sonar images collected by sensors, and labeled one by one. Next, the convolutional neural network model was trained by the set on the workstation platform. The trained model was integrated into the field-programmable gate array system and applied to recognize actual datasets. The recognition results were compared with those of the workstation platform. The comparison shows that the computational precision of the designed field-programmable gate array system based on convolutional neural network is equivalent to that of the workstation platform; however, the recognition time of the designed system can be saved by more than 77%, and its energy consumption can also be saved by more than 96.67%. Therefore, our system basically satisfies our demand for energy-efficient, real-time, and accurate recognition of sonar images.

A Real-Time Deep Learning OFDM Receiver

2022 ◽  
Vol 15 (3) ◽  
pp. 1-25
Author(s):  
Stefan Brennsteiner ◽  
Tughrul Arslan ◽  
John Thompson ◽  
Andrew McCormick
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Real Time ◽  
Field Programmable Gate Array ◽  
Orthogonal Frequency Division Multiplexing ◽  
Frequency Division Multiplexing ◽  
Frequency Division ◽  
Field Programmable ◽  
Gate Array ◽  
Fully Connected

Machine learning in the physical layer of communication systems holds the potential to improve performance and simplify design methodology. Many algorithms have been proposed; however, the model complexity is often unfeasible for real-time deployment. The real-time processing capability of these systems has not been proven yet. In this work, we propose a novel, less complex, fully connected neural network to perform channel estimation and signal detection in an orthogonal frequency division multiplexing system. The memory requirement, which is often the bottleneck for fully connected neural networks, is reduced by ≈ 27 times by applying known compression techniques in a three-step training process. Extensive experiments were performed for pruning and quantizing the weights of the neural network detector. Additionally, Huffman encoding was used on the weights to further reduce memory requirements. Based on this approach, we propose the first field-programmable gate array based, real-time capable neural network accelerator, specifically designed to accelerate the orthogonal frequency division multiplexing detector workload. The accelerator is synthesized for a Xilinx RFSoC field-programmable gate array, uses small-batch processing to increase throughput, efficiently supports branching neural networks, and implements superscalar Huffman decoders.

Sign in / Sign up

Export Citation Format

Share Document