Project 1 (Mini Project on Vehicle Direction Detection

      Vehicle Direction Detection 

          Vehicle detection and statistics in highway monitoring video scenes are of considerable significance to intelligent traffic management and control of the highway. With the popular installation of traffic surveillance cameras, a vast database of traffic video footage has been obtained for analysis. Generally, at a high viewing angle, a more-distant road surface can be considered. The object size of the vehicle changes greatly at this viewing angle, and the detection accuracy of a small object far away from the road is low. In the face of complex camera scenes, it is essential to effectively solve the above problems and further apply them. In this article, we focus on the above issues to propose a viable solution, and we apply the vehicle detection results to multi-object tracking and vehicle counting.Camera reads the road signs & stores into its memory with unique values for left turn, right turn & straight drive. Depending on the direction it is taking, final indication is given to the driver – as an indication if he is driving in the recommended direction or not.Vehicle Direction Determination can also be coupled along - side features like GPS systems to identify whether the vehicle is reaching its destination in an optimized manner. This sub feature can also be used along with Lane Detection, Highway Warning, Ramp Entry / Exit in Wrong Way Detection etc.

At present, vision-based vehicle object detection is divided into traditional machine vision methods and complex deep learning methods. Traditional machine vision methods use the motion of a vehicle to separate it from a fixed background image. This method can be divided into three categories the method of using background subtraction, the method of using continuous video frame difference , and the method of using optical flow. Using the video frame difference method, the variance is calculated according to the pixel values of two or three consecutive video frames. Moreover, the moving foreground region is separated by the threshold . By using this method and suppressing noise, the stopping of the vehicle can also be detected . When the background image in the video is fixed, the background information is used to establish the background model . Then, each frame image is compared with the background model, and the moving object can also be segmented. The method of using optical flow can detect the motion region in the video. The generated optical flow field represents each pixel’s direction of motion and pixel speed . Vehicle detection methods using vehicle features, such as the Scale Invariant Feature Transform (SIFT) and Speeded Up Robust Features (SURF) methods, have been widely used. For example, 3D models have been used to complete vehicle detection and classification tasks . Using the correlation curves of 3D ridges on the outer surface of the vehicle, the vehicles are divided into three categories: cars, SUVs, and minibuses.

The use of deep convolutional networks (CNNs) has achieved amazing success in the field of vehicle object detection. CNNs have a strong ability to learn image features and can perform multiple related tasks, such as classification and bounding box regression . The detection method can be generally divided into two categories. The two-stage method generates a candidate box of the object via various algorithms and then classifies the object by a convolutional neural network. The one-stage method does not generate a candidate box but directly converts the positioning problem of the object bounding box into a regression problem for processing. In the two-stage method, Region-CNN (R-CNN)  uses selective region search  in the image. The image input to the convolutional network must be fixed-size, and the deeper structure of the network requires a long training time and consumes a large amount of storage memory. Drawing on the idea of spatial pyramid matching, SPP NET  allows the network to input images of various sizes and to have fixed outputs. R-FCN, FPN, and Mask RCNN have improved the feature extraction methods, feature selection, and classification capabilities of convolutional networks in different ways. Among the one-stage methods, the most important are the Single Shot Multibox Detector (SSD)  and You Only Look Once (YOLO)  frameworks. The MutiBox , Region Proposal Network (RPN) and multi-scale representation methods are used in SSD, which uses a default set of anchor boxes with different aspect ratios to more accurately position the object. Unlike SSD, the YOLO network divides the image into a fixed number of grids. Each grid is responsible for predicting objects whose centre points are within the grid. YOLOv2  added the BN (Batch Normalization) layer, which makes the network normalize the input of each layer and accelerate the network convergence speed. YOLOv2 uses a multi-scale training method to randomly select a new image size for every ten batches. Our vehicle object detection uses the YOLOv3  network. Based on YOLOv2, YOLOv3 uses logistic regression for the object category. The category loss method is two-class cross-entropy loss, which can handle multiple label problems for the same object. Moreover, logistic regression is used to regress the box confidence to determine if the IOU of the a priori box and the actual box is greater than 0.5. If more than one priority box satisfies the condition, only the largest prior box of the IOU is taken. In the final object prediction, YOLOv3 uses three different scales to predict the object in the image.

              The traditional machine vision method has a faster speed when detecting the vehicle but does not produce a good result when the image changes in brightness, there is periodic motion in the background, and where there are slow moving vehicles or complex scenes. Advanced CNN has achieved good results in object detection; however, CNN is sensitive to scale changes in object detection . The one stage method uses grids to predict objects, and the grid’s spatial constraints make it impossible to have higher precision with the two-stage approach, especially for small objects. The two stage method uses region of interest pooling to segment candidate regions into blocks according to given parameters, and if the candidate region is smaller than the size of the given parameters, the candidate region is padded to the size of the given parameters. In this way, the characteristic structure of a small object is destroyed and its detection accuracy is low. The existing methods do not distinguish if large and small objects belong to the same category. The same method is used to deal with the same type of object, which will also lead to inaccurate detection. The use of image pyramids or multi-scale input images can solve the above problems, although the calculation requirements are large.

Procedure:

1. Create 2 subsytem as per the given instruction in the requirment document.

2.Creating simulink data dictionary (SLDD)

3.Generating a model advisor report

4. Generation of code.

Requirement - 1:

• Steering wheel input as yaw rate (Signal name: SteeringWheel_YawDegreeInput) is the input for this system.
• This is compared against 3 angular values, one each for left turn, right turn & straight drive (Calibration Values: Right_Turn_AngularLimit, Left_Turn_AngularLimit, Straight_Drive_Steering_Angle) to say which specific direction the steering wheel is turning towards.
• Use switch blocks to compare & develop this requirement. Keep this requirement in a subsystem & output of this requirement is a local signal (Signal Name: Vehicle_Turn_Status).
• Here the switch has the threshold values as >=30 and ~=0 as shown below.

Requirement – 2:

• Keep this requirement as a separate subsystem. Inputs to this requirement are local signal from requirement 1 (Signal Name: Vehicle_Turn_Status) & an input signal from camera (Signal Name: CameraInput_RoadSign), which confirms the occurrence of a road sign.
• Signal Vehicle_Turn_Status is compared against calibration values (Calibration Values: RightTurn_RoadSign, LeftTurn_RoadSign, Straight_RoadSign), if each of them is found equal, then each of the three corresponding output is compared against the camera input signal,
• Using a logical operator block, only one among them is finally given as output signal (Signal Name: Vehicle_Direction_Indicator).

Linking the SLDD with the model:

 Sldd data are here in signal and parameter form:

Code generation (C-code):