Month: February 2024

Have you ever pondered the number of vehicles present in a traffic scenario? Have you ever considered the volume of cars navigating through dense traffic? Have you ever contemplated the impact of such traffic on the surrounding routes? We’ve shared these same curiosities and realized the potential to analyze them using computer vision algorithms. We utilized a series of time-lapse images using the coco dataset capturing traffic over a duration and applied an object detection algorithm using YOLO: Real-Time Object Detection to determine the count of vehicles passing within that timeframe.

The graph was plotted with the number of vehicles on the x-axis against the time taken on the y-axis.The depicted graph illustrates the quantity of vehicles captured by the camera within specific time intervals, condensed from hours to mere seconds. Analyzing this graph reveals a substantial and progressively increasing volume of vehicles at each time interval.

It’s quite remarkable to observe how traffic gradually accumulates over time. Once it initiates at a certain point, it persists until there are no more vehicles present. The buildup of traffic is relentless, ceasing only when either all vehicles have cleared or only a few remain. What’s even more intriguing is our ability to analyze this phenomenon using solely an object detection model.

Numerous occurrences lend themselves to analysis through the application of an object detection model. For instance, we could employ it to scrutinize events like a total solar eclipse, assessing the percentage of the sun obscured during specific time intervals. However, implementation was not feasible as the model lacked training for astronomical events or the analysis of celestial bodies, including those in distant space.

This blog was created by Parithosh Dharmapalan and Pooja Srinivasan.

Since the dawn of time, humanity has always been obsessed with two things: First up is their appearances. With the increased boom in social media, photo editing tools and tweaking images have been at an all time high. The other is dogs. So we decided why not combine the two. 

To start with we chose two dog breeds which are very similar in appearance to see if the classifier can differentiate the two. As you can see the classifier does a very good job in differentiating the Italian Greyhound and the Whippet.

We used three major corruption techniques on our images: editing, doodling and superimposing the images. 

Editing:

The first method we used is editing, where we changed the saturation, exposure, contrast and sharpness of the image. 

After altering the image’s exposure, contrast, saturation, and sharpness, the accuracy of the image classifier in identifying the dog breed decreased.

Overlapping:

Since we noticed that even though the classifier was a bit confused with the edited images, we decided to combine the two images as well as overlap them to see how well it could identify them. And the results are as below:

Doodling:

We then proceeded to doodle on the image by creating a “santa dog” and outlining certain features, the classifier had a hard time to identify the images especially when the different colors were taken into consideration. 

We think the classifier had a hard time classifying  the two images mainly due to the fact that we did choose very similar dog breeds that it was initially hard for us to differentiate between the two and then proceeded to make the situation worse for the classifier with the noise. But overall the classifier did a better job than most humans would have.

Pooja Srinivasan and Parithosh Dharmapalan authored this blog