Initial commit

2024-10-18 18:18:36 -05:00 · 2024-10-18 18:18:36 -05:00 · e8ce9f226c
commit e8ce9f226c
7 changed files with 885 additions and 0 deletions
--- a/.gitignore
+++ b/.gitignore
@ -0,0 +1,15 @@
 ### CMake template
 CMakeLists.txt.user
 CMakeCache.txt
 CMakeFiles
 CMakeScripts
 Testing
 Makefile
 cmake_install.cmake
 install_manifest.txt
 compile_commands.json
 CTestTestfile.cmake
 _deps
 CMakeUserPresets.json
 .idea
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@ -0,0 +1,15 @@
 cmake_minimum_required(VERSION 3.29)
 project(pumpkin_tracker)
 set(CMAKE_CXX_STANDARD 20)
 find_package(OpenCV REQUIRED)
 include_directories(${OpenCV_INCLUDE_DIRS})
 FIND_LIBRARY(DarkHelp darkhelp)
 FIND_LIBRARY(Darknet darknet)
 ADD_EXECUTABLE(pumpkin_tracker main.cpp)
 TARGET_LINK_LIBRARIES(pumpkin_tracker ${OpenCV_LIBS})
 TARGET_LINK_LIBRARIES(pumpkin_tracker /usr/local/lib/libdarkhelp.so)
 TARGET_LINK_LIBRARIES(pumpkin_tracker /usr/local/lib/libdarknet.so)
--- a/README.md
+++ b/README.md
@ -0,0 +1,7 @@
 # pumpkin-tracker
 This uses a camera and object detection to estimate the velocity of a flying pumpkin.
 ## Compiling
 Make sure to compile [Darknet](https://github.com/hank-ai/darknet) (the actual neural network),
 [DarkHelp](https://github.com/stephanecharette/DarkHelp) (a C++ API wrapper for Darknet), and install 
 [OpenCV](https://opencv.com) before compiling this project.
--- a/cfg/coco.names
+++ b/cfg/coco.names
@ -0,0 +1,80 @@
 person
 bicycle
 car
 motorbike
 aeroplane
 bus
 train
 truck
 boat
 traffic light
 fire hydrant
 stop sign
 parking meter
 bench
 bird
 cat
 dog
 horse
 sheep
 cow
 elephant
 bear
 zebra
 giraffe
 backpack
 umbrella
 handbag
 tie
 suitcase
 frisbee
 skis
 snowboard
 sports ball
 kite
 baseball bat
 baseball glove
 skateboard
 surfboard
 tennis racket
 bottle
 wine glass
 cup
 fork
 knife
 spoon
 bowl
 banana
 apple
 sandwich
 orange
 broccoli
 carrot
 hot dog
 pizza
 donut
 cake
 chair
 sofa
 pottedplant
 bed
 diningtable
 toilet
 tvmonitor
 laptop
 mouse
 remote
 keyboard
 cell phone
 microwave
 oven
 toaster
 sink
 refrigerator
 book
 clock
 vase
 scissors
 teddy bear
 hair drier
 toothbrush
--- a/cfg/yolov7-tiny.cfg
+++ b/cfg/yolov7-tiny.cfg
@ -0,0 +1,706 @@
 [net]
 # Testing
 #batch=1
 #subdivisions=1
 # Training
 batch=1
 subdivisions=1
 width=416
 height=416
 channels=3
 momentum=0.9
 decay=0.0005
 angle=0
 saturation = 1.5
 exposure = 1.5
 hue=.1
 learning_rate=0.00261
 burn_in=1000
 max_batches = 2000200
 policy=steps
 steps=1600000,1800000
 scales=.1,.1
 # 0
 [convolutional]
 batch_normalize=1
 filters=32
 size=3
 stride=2
 pad=1
 activation=leaky
 # 1
 [convolutional]
 batch_normalize=1
 filters=64
 size=3
 stride=2
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=32
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=32
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=32
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=32
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 8
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [maxpool]
 size=2
 stride=2
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=64
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=64
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 16
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [maxpool]
 size=2
 stride=2
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=128
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=128
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 24
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 [maxpool]
 size=2
 stride=2
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=256
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=256
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 32
 [convolutional]
 batch_normalize=1
 filters=512
 size=1
 stride=1
 pad=1
 activation=leaky
 ##################################
 ### SPPCSP ###
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -2
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 ### SPP ###
 [maxpool]
 stride=1
 size=5
 [route]
 layers=-2
 [maxpool]
 stride=1
 size=9
 [route]
 layers=-4
 [maxpool]
 stride=1
 size=13
 [route]
 layers=-1,-3,-5,-6
 ### End SPP ###
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -10,-1
 # 44
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 ### End SPPCSP ###
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [upsample]
 stride=2
 [route]
 layers = 24
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -1,-3
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=64
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=64
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 56
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [upsample]
 stride=2
 [route]
 layers = 16
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -1,-3
 [convolutional]
 batch_normalize=1
 filters=32
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=32
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=32
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=32
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 68
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 ##########################
 [convolutional]
 batch_normalize=1
 size=3
 stride=2
 pad=1
 filters=128
 activation=leaky
 [route]
 layers = -1,56
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=64
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=64
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=64
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 77
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 size=3
 stride=2
 pad=1
 filters=256
 activation=leaky
 [route]
 layers = -1,44
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [route]
 layers=-2
 [convolutional]
 batch_normalize=1
 filters=128
 size=1
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=128
 size=3
 stride=1
 pad=1
 activation=leaky
 [convolutional]
 batch_normalize=1
 filters=128
 size=3
 stride=1
 pad=1
 activation=leaky
 [route]
 layers = -5,-3,-2,-1
 # 86
 [convolutional]
 batch_normalize=1
 filters=256
 size=1
 stride=1
 pad=1
 activation=leaky
 #############################
 # ============ End of Neck ============ #
 # ============ Head ============ #
 # P3
 [route]
 layers = 68
 [convolutional]
 batch_normalize=1
 size=3
 stride=1
 pad=1
 filters=128
 activation=leaky
 [convolutional]
 size=1
 stride=1
 pad=1
 filters=255
 #activation=linear
 activation=logistic
 [yolo]
 mask = 0,1,2
 anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326
 classes=80
 num=9
 jitter=.1
 scale_x_y = 2.0
 objectness_smooth=1
 ignore_thresh = .7
 truth_thresh = 1
 #random=1
 resize=1.5
 iou_thresh=0.2
 iou_normalizer=0.05
 cls_normalizer=0.5
 obj_normalizer=1.0
 iou_loss=ciou
 nms_kind=diounms
 beta_nms=0.6
 new_coords=1
 max_delta=2
 # P4
 [route]
 layers = 77
 [convolutional]
 batch_normalize=1
 size=3
 stride=1
 pad=1
 filters=256
 activation=leaky
 [convolutional]
 size=1
 stride=1
 pad=1
 filters=255
 #activation=linear
 activation=logistic
 [yolo]
 mask = 3,4,5
 anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326
 classes=80
 num=9
 jitter=.1
 scale_x_y = 2.0
 objectness_smooth=1
 ignore_thresh = .7
 truth_thresh = 1
 #random=1
 resize=1.5
 iou_thresh=0.2
 iou_normalizer=0.05
 cls_normalizer=0.5
 obj_normalizer=1.0
 iou_loss=ciou
 nms_kind=diounms
 beta_nms=0.6
 new_coords=1
 max_delta=2
 # P5
 [route]
 layers = 86
 [convolutional]
 batch_normalize=1
 size=3
 stride=1
 pad=1
 filters=512
 activation=leaky
 [convolutional]
 size=1
 stride=1
 pad=1
 filters=255
 #activation=linear
 activation=logistic
 [yolo]
 mask = 6,7,8
 anchors = 10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326
 classes=80
 num=9
 jitter=.1
 scale_x_y = 2.0
 objectness_smooth=1
 ignore_thresh = .7
 truth_thresh = 1
 #random=1
 resize=1.5
 iou_thresh=0.2
 iou_normalizer=0.05
 cls_normalizer=0.5
 obj_normalizer=1.0
 iou_loss=ciou
 nms_kind=diounms
 beta_nms=0.6
 new_coords=1
 max_delta=2
--- a/cfg/yolov7-tiny.weights
+++ b/cfg/yolov7-tiny.weights
--- a/main.cpp
+++ b/main.cpp
@ -0,0 +1,62 @@
 /* Computer vision demo. Performs object tracking on pumpkins to determine velocity, launch angle, and predicted range.
 *
 */
 #include <iostream>
 #include <DarkHelp.hpp>
 #include <unistd.h>
 #include <opencv4/opencv2/core/mat.hpp>
 #include <opencv4/opencv2/videoio.hpp>
 #include <opencv4/opencv2/highgui.hpp>
 // Default configuration values. Using YOLOv7 tiny model. Make sure to download the weights.
 #define DARKNET_CFG "cfg/yolov7-tiny.cfg"
 #define DARKNET_WEIGHTS "cfg/yolov7-tiny.weights"
 #define DARKNET_DATA "cfg/coco.names"
 #define CAMERA_ID 0
 #define SHOW_GUI true
 using namespace std;
 using namespace cv;
 using namespace DarkHelp;
 int main() {
    // Configure the neural network
    NN nn(DARKNET_CFG, DARKNET_WEIGHTS, DARKNET_DATA);
    // Setup the capture stream from the webcam
    VideoCapture cap(CAMERA_ID);
    if (not cap.isOpened())
    {
        throw std::runtime_error("failed to open the webcam");
    }
    cap.set(CAP_PROP_FRAME_WIDTH, 640.0);
    cap.set(CAP_PROP_FRAME_HEIGHT, 480.0);
    cap.set(CAP_PROP_FPS, 30.0);
    while (cap.isOpened()) {
        Mat frame;
        cap >> frame;
        if (frame.empty()) break;
        const PredictionResults results = nn.predict(frame);
        // print results to console
        cout << "RESULTS" << endl;
        for (int i = 0; i < results.size(); i++) {
            const PredictionResult result = results[i];
            cout << result.name << " | (" << result.original_point.x << ", " << result.original_point.y << ")" << endl;
        }
        cout << endl;
        // Sleep between frames; if the UI is enabled, show the latest (annotated) frame
        if constexpr (SHOW_GUI) {
            frame = nn.annotate();
            imshow("pumpkin-tracker", frame);
            if (const auto key = waitKey(15); key == 27) break;
        }
        else {
            usleep(15000);
        }
    }
 }