Merge pull request #31 from wml-frc/CJ-Patch

Updated readme for raspbian

Author: CJBuchel

README.md

@@ -101,41 +101,64 @@ After the basics are complete you can use some of the provided processing types
 
 CJ::VisionTracking vision;
 
+double offsetX, offsetY;
 int ResWidth = 640, ResHeight = 480;
 
 double cx, cy;
 
 cv::Mat Image; // Origin Image
-cv::Mat ProcessingOutput; // Image in processing
 cv::Mat TrackingImage; // Imaged After it has been procesed
+cv::Mat ProcessingOutput;
 
 void curtin_frc_vision::run() {
+	auto inst = nt::NetworkTableInstance::GetDefault();
+	auto visionTable = inst.GetTable("VisionTracking");
+	auto table = visionTable->GetSubTable("Target");
 
-	vision.SetupVision(&Image, 1, 60, ResHeight, ResWidth, 30, "TestCam", false);
-	vision.CustomTrack(&TrackingImage, &Image, 30, 70, 50, 255, 50, 2, 2);
-	vision.Processing.visionHullGeneration.BoundingBox(&TrackingImage, &ProcessingOutput, &cx, &cy, 10);
+	TargetX = table->GetEntry("Target_X");
+	TargetY = table->GetEntry("Target_Y");
+	ImageHeight = table->GetEntry("ImageHeight");
+	ImageWidth = table->GetEntry("ImageWidth");
 
-  #ifdef __DESKTOP__ 
+	inst.StartClientTeam(4788);
+
+	vision.SetupVision(&Image, 0, 30, ResHeight, ResWidth, 1, "Turret Cam", true);
+	vision.CustomTrack(&TrackingImage, &Image, 30, 70, 0, 255, 0, 0);
+	vision.Processing.visionHullGeneration.BoundingBox(&TrackingImage, &ProcessingOutput, &cx, &cy, 10);
+	#ifdef __DESKTOP__ 
 	std::cout << "Exposure Might be dissabled on local machine" << std::endl;
 	#else
-	system("v4l2-ctl -d /dev/video4 --set-ctrl=exposure_absolute=1");
+	system("v4l2-ctl -d /dev/video0 --set-ctrl=exposure_absolute=1");
 	#endif
 
+	std::cout << "Vision Tracking Process Running" << std::endl;
 	while (true) {
 		if (vision.Camera.cam.sink.GrabFrame(Image) != 0) {
+
 			// Vision Outputing
 			#ifdef __DESKTOP__
 			vision.Output.Display("Origin Image", &Image);
 			vision.Output.Display("Green Filtered Image", &TrackingImage);
 			vision.Output.Display("Contour Detection", &ProcessingOutput);
 			#else 
-			vision.Camera.cam.output.PutFrame(Image);
+			vision.Camera.cam.output.PutFrame(ProcessingOutput); // Might get rid of this to speed up loop.
 			#endif
+
+			//Calc offset
+			offsetX = cx-(ResWidth/2);
+			offsetY = cy-(ResHeight/2);
+
+			TargetX.SetDouble(offsetX);
+			TargetY.SetDouble(offsetY);
+			ImageHeight.SetDouble(ResHeight);
+			ImageWidth.SetDouble(ResWidth);
+		} else {
+			visionTable->PutBoolean("Vision Active", false);
 		}
 	}
 }
 ```
-The above tracks green pixles at a regular exposure then detects and draws bounding boxes on everything that is green. It then sends the center values through network tables. So you can use it on your roborio and shuffleboard for autonomous.
+The above tracks green pixles at a low exposure then detects and draws bounding boxes on everything that is green. It then sends the center values through network tables. So you can use it on your roborio and shuffleboard for autonomous/teleop.
 
 
 # Getting started with your coprocessor
@@ -144,6 +167,7 @@ The above tracks green pixles at a regular exposure then detects and draws bound
   - [Raspberry Pi](https://downloads.raspberrypi.org/raspbian_lite_latest)
 ### 2. Flash the image onto a microsd card using etcher: https://etcher.io/
   - You will need to grab the .img file from the zip. The .img file is what you give to etcher.
+	- If you want a headless setup. (Only power and ethernet) By Default since 2016, rasbian locks ssh when you first boot onto it. You can unlock ssh by placing a file called ssh with no file extension in the accessible partition of the sd. When the board boots up it should search for this file and allow ssh.
   - After complete, insert the microSD card and startup the coprocessor. It may take a minute or two
 ### 3. SSH into the Coprocessor
   - On Mac/Linux, use `ssh username@hostname`