Pinned Repositories
3D-LD-ECG
3D-sensing-Eskin
academicpages.github.io
Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes
cnn_open
A hardware implementation of CNN, written by Verilog and synthesized on FPGA
Convolutional-Neural-Network
Implementation of CNN using Verilog
ECG-SIGNAL-Pre-Processing-Using-NLMS-LMS-Adaptive-Filters
In this project, the LMS and NLMS adaptive filters are implemented to remove the baseline wander, EMG and motion artifacts from a given signal and compare the results to deduce important distinctions regarding the respective performances.
Fpga-Implementation-of-Precise-Convolutional-Neural-Network-for-Extreme-Learning-Machine
Feed-forward neural networks can be trained based on a gradient-descent based backpropagation algorithm. But, these algorithms require more computation time. Extreme Learning Machines (ELM’s) are time-efficient, and they are less complicated than the conventional gradient-based algorithm. In previous years, an SRAM based convolutional neural network using a receptive – field Approach was proposed. This neural network was used as an encoder for the ELM algorithm and was implemented on FPGA. But, this neural network used an inaccurate 3-stage pipelined parallel adder. Hence, this neural network generates imprecise stimuli to the hidden layer neurons. This paper presents an implementation of precise convolutional neural network for encoding in the ELM algorithm based on the receptive - field approach at the hardware level. In the third stage of the pipelined parallel adder, instead of approximating the output by using one 2-input 15-bit adder, one 4-input 14-bit adder is used. Also, an additional weighted pixel array block is used. This weighted pixel array improves the accuracy of generating 128 weighted pixels. This neural network was simulated using ModelSim-Altera 10.1d and synthesized using Quartus II 13.0 sp1. This neural network is implemented on Cyclone V FPGA and used for pattern recognition applications. Although this design consumes slightly more hardware resources, this design is more accurate compared to previously existing encoders.
Graph-RAG
A graph rag for PDFs based on langchain and Neo4j. Can fetch PDFs from Zotero Library through zotero api.
stereopsis-anything
Stereo Anything can convert 2D content on the screen in real-time into stereoscopic images (spatial videos) that are theoretically compatible with various AR/VR glasses, such as Rayneo Air 1s/2s, X1, X2, Nreal Air, etc.
WeTac
zjkhurry's Repositories
zjkhurry/stereopsis-anything
Stereo Anything can convert 2D content on the screen in real-time into stereoscopic images (spatial videos) that are theoretically compatible with various AR/VR glasses, such as Rayneo Air 1s/2s, X1, X2, Nreal Air, etc.
zjkhurry/WeTac
zjkhurry/Graph-RAG
A graph rag for PDFs based on langchain and Neo4j. Can fetch PDFs from Zotero Library through zotero api.
zjkhurry/3D-LD-ECG
zjkhurry/3D-sensing-Eskin
zjkhurry/academicpages.github.io
Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes
zjkhurry/cnn_open
A hardware implementation of CNN, written by Verilog and synthesized on FPGA
zjkhurry/Convolutional-Neural-Network
Implementation of CNN using Verilog
zjkhurry/ECG-SIGNAL-Pre-Processing-Using-NLMS-LMS-Adaptive-Filters
In this project, the LMS and NLMS adaptive filters are implemented to remove the baseline wander, EMG and motion artifacts from a given signal and compare the results to deduce important distinctions regarding the respective performances.
zjkhurry/Fpga-Implementation-of-Precise-Convolutional-Neural-Network-for-Extreme-Learning-Machine
Feed-forward neural networks can be trained based on a gradient-descent based backpropagation algorithm. But, these algorithms require more computation time. Extreme Learning Machines (ELM’s) are time-efficient, and they are less complicated than the conventional gradient-based algorithm. In previous years, an SRAM based convolutional neural network using a receptive – field Approach was proposed. This neural network was used as an encoder for the ELM algorithm and was implemented on FPGA. But, this neural network used an inaccurate 3-stage pipelined parallel adder. Hence, this neural network generates imprecise stimuli to the hidden layer neurons. This paper presents an implementation of precise convolutional neural network for encoding in the ELM algorithm based on the receptive - field approach at the hardware level. In the third stage of the pipelined parallel adder, instead of approximating the output by using one 2-input 15-bit adder, one 4-input 14-bit adder is used. Also, an additional weighted pixel array block is used. This weighted pixel array improves the accuracy of generating 128 weighted pixels. This neural network was simulated using ModelSim-Altera 10.1d and synthesized using Quartus II 13.0 sp1. This neural network is implemented on Cyclone V FPGA and used for pattern recognition applications. Although this design consumes slightly more hardware resources, this design is more accurate compared to previously existing encoders.
zjkhurry/fpga_image_processing
IP operations in verilog (simulation and implementation on ice40)
zjkhurry/HelloWord-Keyboard
zjkhurry/OSlw_Code
Code for OSLW
zjkhurry/PyXA
Python for Automation
zjkhurry/UPduino-v3.0
UPduino 3.0: new 4 layer layout, various other improvements