Lane Detection Optimized with GPU Coder

This example shows how to generate CUDA® code from a deep learning network, represented by a SeriesNetwork object. In this example, the series network is a convolutional neural network that can detect and output lane marker boundaries from an image.


  • CUDA enabled NVIDIA® GPU with compute capability 3.2 or higher.

  • NVIDIA CUDA toolkit and driver.

  • NVIDIA cuDNN library.

  • OpenCV libraries for video read and image display operations.

  • Environment variables for the compilers and libraries. For information on the supported versions of the compilers and libraries, see Third-party Products (GPU Coder). For setting up the environment variables, see Setting Up the Prerequisite Products (GPU Coder).

  • GPU Coder Interface for Deep Learning Libraries support package. To install this support package, use the Add-On Explorer.

Verify GPU Environment

Use the coder.checkGpuInstall function to verify that the compilers and libraries necessary for running this example are set up correctly.

envCfg = coder.gpuEnvConfig('host');
envCfg.DeepLibTarget = 'cudnn';
envCfg.DeepCodegen = 1;
envCfg.Quiet = 1;

Get Pretrained SeriesNetwork

[laneNet, coeffMeans, coeffStds] = getLaneDetectionNetwork();

This network takes an image as an input and outputs two lane boundaries that correspond to the left and right lanes of the ego vehicle. Each lane boundary is represented by the parabolic equation: , where y is the lateral offset and x is the longitudinal distance from the vehicle. The network outputs the three parameters a, b, and c per lane. The network architecture is similar to AlexNet except that the last few layers are replaced by a smaller fully connected layer and regression output layer.

ans = 

  23x1 Layer array with layers:

     1   'data'          Image Input                   227x227x3 images with 'zerocenter' normalization
     2   'conv1'         Convolution                   96 11x11x3 convolutions with stride [4  4] and padding [0  0  0  0]
     3   'relu1'         ReLU                          ReLU
     4   'norm1'         Cross Channel Normalization   cross channel normalization with 5 channels per element
     5   'pool1'         Max Pooling                   3x3 max pooling with stride [2  2] and padding [0  0  0  0]
     6   'conv2'         Convolution                   256 5x5x48 convolutions with stride [1  1] and padding [2  2  2  2]
     7   'relu2'         ReLU                          ReLU
     8   'norm2'         Cross Channel Normalization   cross channel normalization with 5 channels per element
     9   'pool2'         Max Pooling                   3x3 max pooling with stride [2  2] and padding [0  0  0  0]
    10   'conv3'         Convolution                   384 3x3x256 convolutions with stride [1  1] and padding [1  1  1  1]
    11   'relu3'         ReLU                          ReLU
    12   'conv4'         Convolution                   384 3x3x192 convolutions with stride [1  1] and padding [1  1  1  1]
    13   'relu4'         ReLU                          ReLU
    14   'conv5'         Convolution                   256 3x3x192 convolutions with stride [1  1] and padding [1  1  1  1]
    15   'relu5'         ReLU                          ReLU
    16   'pool5'         Max Pooling                   3x3 max pooling with stride [2  2] and padding [0  0  0  0]
    17   'fc6'           Fully Connected               4096 fully connected layer
    18   'relu6'         ReLU                          ReLU
    19   'drop6'         Dropout                       50% dropout
    20   'fcLane1'       Fully Connected               16 fully connected layer
    21   'fcLane1Relu'   ReLU                          ReLU
    22   'fcLane2'       Fully Connected               6 fully connected layer
    23   'output'        Regression Output             mean-squared-error with 'leftLane_a', 'leftLane_b', and 4 other responses

Examine Main Entry-Point Function

type detect_lane.m
function [laneFound, ltPts, rtPts] = detect_lane(frame, laneCoeffMeans, laneCoeffStds) 
% From the networks output, compute left and right lane points in the 
% image coordinates. The camera coordinates are described by the caltech 
% mono camera model.


% A persistent object mynet is used to load the series network object.
% At the first call to this function, the persistent object is constructed and
% setup. When the function is called subsequent times, the same object is reused 
% to call predict on inputs, thus avoiding reconstructing and reloading the
% network object.
persistent lanenet;

if isempty(lanenet)
    lanenet = coder.loadDeepLearningNetwork('laneNet.mat', 'lanenet');

lanecoeffsNetworkOutput = lanenet.predict(permute(frame, [2 1 3]));

% Recover original coeffs by reversing the normalization steps

params = lanecoeffsNetworkOutput .* laneCoeffStds + laneCoeffMeans;

isRightLaneFound = abs(params(6)) > 0.5; %c should be more than 0.5 for it to be a right lane
isLeftLaneFound =  abs(params(3)) > 0.5;

vehicleXPoints = 3:30; %meters, ahead of the sensor
ltPts = coder.nullcopy(zeros(28,2,'single'));
rtPts = coder.nullcopy(zeros(28,2,'single'));

if isRightLaneFound && isLeftLaneFound
    rtBoundary = params(4:6);		
	rt_y = computeBoundaryModel(rtBoundary, vehicleXPoints);
	ltBoundary = params(1:3);
	lt_y = computeBoundaryModel(ltBoundary, vehicleXPoints);
	% Visualize lane boundaries of the ego vehicle
    tform = get_tformToImage;
    % map vehicle to image coordinates
    ltPts =  tform.transformPointsInverse([vehicleXPoints', lt_y']);
    rtPts =  tform.transformPointsInverse([vehicleXPoints', rt_y']);
	laneFound = true;
	laneFound = false;


function yWorld = computeBoundaryModel(model, xWorld)
	yWorld = polyval(model, xWorld);	

function tform = get_tformToImage 
% Compute extrinsics based on camera setup
yaw = 0;
pitch = 14; % pitch of the camera in degrees
roll = 0;

translation = translationVector(yaw, pitch, roll);
rotation    = rotationMatrix(yaw, pitch, roll);

% Construct a camera matrix
focalLength    = [309.4362, 344.2161];
principalPoint = [318.9034, 257.5352];
Skew = 0;

camMatrix = [rotation; translation] * intrinsicMatrix(focalLength, ...
	Skew, principalPoint);

% Turn camMatrix into 2-D homography
tform2D = [camMatrix(1,:); camMatrix(2,:); camMatrix(4,:)]; % drop Z

tform = projective2d(tform2D);
tform = tform.invert();

function translation = translationVector(yaw, pitch, roll)
SensorLocation = [0 0];
Height = 2.1798;    % mounting height in meters from the ground
rotationMatrix = (...
	rotZ(yaw)*... % last rotation
	rotZ(roll)... % first rotation

% Adjust for the SensorLocation by adding a translation
sl = SensorLocation;

translationInWorldUnits = [sl(2), sl(1), Height];
translation = translationInWorldUnits*rotationMatrix;

% Rotation around X-axis
function R = rotX(a)
a = deg2rad(a);
R = [...
	1   0        0;
	0   cos(a)  -sin(a);
	0   sin(a)   cos(a)];


% Rotation around Y-axis
function R = rotY(a)
a = deg2rad(a);
R = [...
	cos(a)  0 sin(a);
	0       1 0;
	-sin(a) 0 cos(a)];


% Rotation around Z-axis
function R = rotZ(a)
a = deg2rad(a);
R = [...
	cos(a) -sin(a) 0;
	sin(a)  cos(a) 0;
	0       0      1];

% Given the Yaw, Pitch, and Roll, determine the appropriate Euler
% angles and the sequence in which they are applied to
% align the camera's coordinate system with the vehicle coordinate
% system. The resulting matrix is a Rotation matrix that together
% with the Translation vector defines the extrinsic parameters of the camera.
function rotation = rotationMatrix(yaw, pitch, roll)

rotation = (...
	rotY(180)*...            % last rotation: point Z up
	rotZ(-90)*...            % X-Y swap
	rotZ(yaw)*...            % point the camera forward
	rotX(90-pitch)*...       % "un-pitch"
	rotZ(roll)...            % 1st rotation: "un-roll"

function intrinsicMat = intrinsicMatrix(FocalLength, Skew, PrincipalPoint)
intrinsicMat = ...
	[FocalLength(1)  , 0                     , 0; ...
	 Skew             , FocalLength(2)   , 0; ...
	 PrincipalPoint(1), PrincipalPoint(2), 1];

Generate Code for Network and Post-Processing Code

The network computes parameters a, b, and c that describe the parabolic equation for the left and right lane boundaries.

From these parameters, compute the x and y coordinates corresponding to the lane positions. The coordinates must be mapped to image coordinates. The function detect_lane.m performs all these computations. Generate CUDA code for this function by creating a GPU code configuration object for a 'lib' target and set the target language to C++. Use the coder.DeepLearningConfig function to create a CuDNN deep learning configuration object and assign it to the DeepLearningConfig property of the GPU code configuration object. Run the codegen command.

cfg = coder.gpuConfig('lib');
cfg.DeepLearningConfig = coder.DeepLearningConfig('cudnn');
cfg.GenerateReport = true;
cfg.TargetLang = 'C++';
codegen -args {ones(227,227,3,'single'),ones(1,6,'double'),ones(1,6,'double')} -config cfg detect_lane
Code generation successful: To view the report, open('codegen/lib/detect_lane/html/report.mldatx').

Generated Code Description

The series network is generated as a C++ class containing an array of 23 layer classes.

class c_lanenet
  int32_T batchSize;
  int32_T numLayers;
  real32_T *inputData;
  real32_T *outputData;
  MWCNNLayer *layers[23];
  void setup(void);
  void predict(void);
  void cleanup(void);

The setup() method of the class sets up handles and allocates memory for each layer object. The predict() method invokes prediction for each of the 23 layers in the network.

The cnn_lanenet_conv*_w and cnn_lanenet_conv*_b files are the binary weights and bias file for convolution layer in the network. The cnn_lanenet_fc*_w and cnn_lanenet_fc*_b files are the binary weights and bias file for fully connected layer in the network.

codegendir = fullfile('codegen', 'lib', 'detect_lane');
.                                      cnn_lanenet_conv5_b                    
..                                     cnn_lanenet_conv5_w                            cnn_lanenet_data_offset                
DeepLearningNetwork.h                  cnn_lanenet_data_scale                 
DeepLearningNetwork.o                  cnn_lanenet_fc6_b                                   cnn_lanenet_fc6_w                      
MWCNNLayerImpl.hpp                     cnn_lanenet_fcLane1_b                  
MWCNNLayerImpl.o                       cnn_lanenet_fcLane1_w                             cnn_lanenet_fcLane2_b                  
MWCudaDimUtility.h                     cnn_lanenet_fcLane2_w                  
MWCudaDimUtility.o                     cnn_lanenet_responseNames.txt          
MWElementwiseAffineLayer.cpp           codeInfo.mat                           
MWElementwiseAffineLayer.hpp           detect_lane.a                          
MWElementwiseAffineLayer.o                           detect_lane.h                          
MWElementwiseAffineLayerImpl.hpp       detect_lane.o                          
MWElementwiseAffineLayerImpl.o             detect_lane_data.h                     
MWElementwiseAffineLayerImplKernel.o   detect_lane_data.o                     
MWFusedConvReLULayer.hpp               detect_lane_initialize.h               
MWFusedConvReLULayer.o                 detect_lane_initialize.o                  detect_lane_ref.rsp                    
MWKernelHeaders.hpp                    detect_lane_rtwutil.h                          detect_lane_rtwutil.o                  
MWTargetNetworkImpl.o                  detect_lane_terminate.h                
buildInfo.mat                          detect_lane_terminate.o                
cnn_api.cpp                            detect_lane_types.h                    
cnn_api.hpp                            examples                               
cnn_api.o                              gpu_codegen_info.mat                   
cnn_lanenet_conv1_b                    html                                   
cnn_lanenet_conv1_w                    interface                              
cnn_lanenet_conv2_w                    predict.h                              
cnn_lanenet_conv3_b                    predict.o                              
cnn_lanenet_conv3_w                    rtw_proj.tmw                           
cnn_lanenet_conv4_b                    rtwtypes.h                             

Generate Additional Files for Post-Processing the Output

Export mean and std values from the trained network for use during execution.

codegendir = fullfile(pwd, 'codegen', 'lib','detect_lane');
fid = fopen(fullfile(codegendir,'mean.bin'), 'w');
A = [coeffMeans coeffStds];
fwrite(fid, A, 'double');

Main File

Compile the network code by using a main file. The main file uses the OpenCV VideoCapture method to read frames from the input video. Each frame is processed and classified until no more frames are read. Before displaying the output for each frame, the outputs are post-processed by using the detect_lane function generated in detect_lane.cpp.

type main_lanenet.cpp
/* Copyright 2016 The MathWorks, Inc. */

#include <stdio.h>
#include <stdlib.h>
#include <cuda.h>
#include "opencv2/opencv.hpp"
#include <list>
#include <cmath>
#include "detect_lane.h"

using namespace cv;
void readData(float *input, Mat& orig, Mat & im)
	Size size(227,227);
	for(int j=0;j<227*227;j++)
		//BGR to RGB

void addLane(float pts[28][2], Mat & im, int numPts)
    std::vector<Point2f> iArray;
    for(int k=0; k<numPts; k++) 
    Mat curve(iArray, true);
    curve.convertTo(curve, CV_32S); //adapt type for polylines
    polylines(im, curve, false, CV_RGB(255,255,0), 2, CV_AA);

void writeData(float *outputBuffer, Mat & im, int N, double means[6], double stds[6])
    // get lane coordinates
    boolean_T laneFound = 0;	
    float ltPts[56];
    float rtPts[56];	
    detect_lane(outputBuffer, means, stds, &laneFound, ltPts, rtPts);    
	if (!laneFound)
	float ltPtsM[28][2];
	float rtPtsM[28][2];
	for(int k=0; k<28; k++)
		ltPtsM[k][0] = ltPts[k];
		ltPtsM[k][1] = ltPts[k+28];
		rtPtsM[k][0] = rtPts[k];
		rtPtsM[k][1] = rtPts[k+28];   

	addLane(ltPtsM, im, 28);
	addLane(rtPtsM, im, 28);

void readMeanAndStds(const char* filename, double means[6], double stds[6])
    FILE* pFile = fopen(filename, "rb");
    if (pFile==NULL)
        fputs ("File error",stderr);

    // obtain file size
    fseek (pFile , 0 , SEEK_END);
    long lSize = ftell(pFile);
    double* buffer = (double*)malloc(lSize);
    size_t result = fread(buffer,sizeof(double),lSize,pFile);
    if (result*sizeof(double) != lSize) {    
        fputs ("Reading error",stderr);
    for (int k = 0 ; k < 6; k++)
        means[k] = buffer[k];
        stds[k] = buffer[k+6];

// Main function
int main(int argc, char* argv[])
    float *inputBuffer = (float*)calloc(sizeof(float),227*227*3);
    float *outputBuffer = (float*)calloc(sizeof(float),6);

    if ((inputBuffer == NULL) || (outputBuffer == NULL)) {
        printf("ERROR: Input/Output buffers could not be allocated!\n");
    // get ground truth mean and std
    double means[6];
    double stds[6];	
    readMeanAndStds("mean.bin", means, stds);	
	if (argc < 2)
        printf("Pass in input video file name as argument\n");
        return -1;
    VideoCapture cap(argv[1]);
    if (!cap.isOpened()) {
        printf("Could not open the video capture device.\n");
        return -1;

    cudaEvent_t start, stop;
    float fps = 0;
    Mat orig, im;    
    namedWindow("Lane detection demo",CV_WINDOW_NORMAL);
        cap >> orig;
        if (orig.empty()) break;                
        readData(inputBuffer, orig, im);		

        writeData(inputBuffer, orig, 6, means, stds);
        char strbuf[50];
        float milliseconds = -1.0; 
        cudaEventElapsedTime(&milliseconds, start, stop);
        fps = fps*.9+1000.0/milliseconds*.1;
        sprintf (strbuf, "%.2f FPS", fps);
        putText(orig, strbuf, cvPoint(200,30), CV_FONT_HERSHEY_DUPLEX, 1, CV_RGB(0,0,0), 2);
        imshow("Lane detection demo", orig); 		
        if( waitKey(50)%256 == 27 ) break; // stop capturing by pressing ESC	*/       
    destroyWindow("Lane detection demo");
    return 0;

Download Example Video

if ~exist('./caltech_cordova1.avi', 'file')
    url = '';
    websave('caltech_cordova1.avi', url);

Build Executable

if ispc
    setenv('MATLAB_ROOT', matlabroot);
    vcvarsall = mex.getCompilerConfigurations('C++').Details.CommandLineShell;
    setenv('VCVARSALL', vcvarsall);
    [~,~] = system('make_win_lane_detection.bat');
    [status,cmdout] = system('lanenet.exe ..\..\..\caltech_cordova1.avi');
    setenv('MATLAB_ROOT', matlabroot);
    [~,~] = system('make -f');
    [status,cmdout] = system('./lanenet ../../../caltech_cordova1.avi');
./lanenet ../../../caltech_cordova1.avi: Segmentation fault

Input Screenshot

Output Screenshot

Related Topics