Man fined $400 for scratching his head while driving due to smart camera error

camera

Man fined $400 for scratching his head while driving due to smart camera error

A Dutchman was fined €380 after an AI camera caught him talking on the phone while driving. True, he claims that he just scratched his head, and the system made a mistake.

Last November, Tim Hansen received a ticket for allegedly talking on a cell phone while driving a month earlier. He was shocked, mainly because he did not remember using his phone while driving that day, and decided to check the incriminating photo on the website of the Central Judicial Recovery Agency. At first glance, it appears that Tim is actually talking on the phone, but upon closer inspection, it turns out that he is not actually holding anything in his hand. The man was simply scratching the back of his head, and the camera took the position of his hand as holding a phone. Even more puzzling was the fact that the person who checked the image and confirmed its correctness also did not notice the “false positive.”

Hansen, who works in information technology and creates algorithms for editing and analyzing images, used his personal experience to explain how the Monocam police camera system works and why it can make mistakes. Although he could not personally test the system, he explained how it works and why it can produce false positives.

“In my case, the AI ​​model showed that I was holding a phone in my hands, although I was not. Then we talk about a false positive. An ideal model would only predict true positives and true negatives, but being 100% correct is rare,” Tim wrote.

As the IT specialist explained, systems like Monocam must be trained on a large set of images, divided into two or three groups: a training set, a validation set and a test set. The first set is used to teach the algorithm what objects are in certain images and what properties (colors, lines, etc.) are inherent in them, the second is used to optimize a number of parameters of the algorithm, and the third is to check how well the system actually works.

“The algorithm we used and the police algorithm would suspect the presence of a phone because in the training dataset there are many examples of people making calls with a phone in their hand near their ear,” Tim said. “It is entirely possible that the training data set contains few or no photographs of people sitting with an empty hand to their ear. In this case, it becomes less important for the algorithm whether the phone is actually in the hand, but it is enough that the hand is next to the ear. To improve the situation, we should add more photographs in which the hand is empty.”

Hansen argues that because of the many variables that can influence an algorithm’s decision, a human filter is needed to minimize the number of false positives. Only in his case the fine was confirmed by a person after analyzing the picture taken by the camera, so this is also not a reliable solution.

The Dutch driver has contested the fine and expects a positive outcome, but will now have to wait up to 26 weeks for an official decision. His case has received widespread publicity in the Netherlands and neighboring countries such as Belgium, where some agencies require the installation of cameras capable of recording the use of mobile phones while driving, but Tim’s story proves that they are far from 100% reliable.

hot LoL Patch 14.6: Riot reveals the next Crystallis…


hot BTS member Suga’s concert film D-Day will be released in…


hot This is why game journalists are “bad” at video games


hot The Gentlemen on Netflix – Should you watch the film…


hot Shogun release schedule: When is episode 4 coming out? |…


hot The new Assassin’s Creed title moves away from the game, clues…


hot The use of AI in board games causes a new scandal


hot Fortnite Chapter 5 Season 2: Mythical weapons, bosses and…


hot Halo Infinite: A new map with Operation Cyber ​​Showdown III


hot New LoL High Noon skins: release date, champions and more



Bradley Gervais

Meet Bradley Gervais, the maestro of Gaming Mods and IT wizardry. With a profound understanding of gaming intricacies, Bradley transforms digital landscapes, pushing the boundaries of gaming experiences through innovative modifications. His expertise extends beyond the virtual realm, seamlessly intertwining with the complexities of Information Technology. Bradley's skill set is a fusion of creativity and technical prowess, earning him a reputation as a go-to authority in the gaming modding community. Whether crafting immersive game enhancements or navigating the IT labyrinth, Bradley Gervais is your guide to a seamlessly blended world where gaming prowess meets technological finesse.