Tag Archives: cucumber


How one Japanese used Deep Learning to sort his parents’ cucumbers

When we think of neural networks or machine learning we tend to imagine them involved in radically transforming the world by preventing crimes before they happen or improving energy efficiency. But you don’t need to change the world to make good use of machine learning. Sometimes, using this tech to help your mom and pop’s business can be just as impressive.


Credit: Pixabay

Makoto Koike, a former embedded systems designer, first got into machine learning after he found out that Google’s DeepMind division used it to train a machine to beat a human champion at Go. He later learned about TensorFlow, an open source library released by Google that allows engineers to implement deep neural networks without prior knowledge of all the complex mathematical and optimization algorithms. Using TensorFlow, Koike saved countless man-hours for his parents who run a cucumber farm in Japan.

What deep learning does is it allows a computer to learn from training data, like a huge database of images, what the most important or defining features that it needs to recognize are. With the help of a hierarchy of numerous artificial neurons, deep learning can then classify all sorts of images with high accuracy. A neural network could be used to recognize different species of cats from varying images or even predict what happens next.

Crooked cucumbers are ranked lower. Credit: Makoto Koike

Crooked cucumbers are ranked lower. Credit: Makoto Koike

His deep learning system uses a Raspberry Pi 3 as the controller and a personal computer that’s connected to the TensorFlow server. Once a cucumber travels down the line and in front of the system’s camera, a picture is taken and then Makoto’s modified neural network tries to ‘understand’ what it’s looking it at. First, it determines that the item sitting right in front of the camera is a cucumber, then it further classifies it because, if you didn’t know, cucumbers come in all sort of sizes, textures, and varying degrees of quality. For instance, thorny cucumbers with many prickles still on them that are fresh and vivid in color are sold at a premium, so sorting the gems from the clutter can be very lucrative.

“The sorting work is not an easy task to learn. You have to look at not only the size and thickness, but also the color, texture, small scratches, whether or not they are crooked and whether they have prickles. It takes months to learn the system and you can’t just hire part-time workers during the busiest period. I myself only recently learned to sort cucumbers well,” Makoto said for Google Cloud’s blog.

Makoto spent about three months taking 7,000 pictures of cucumbers sorted by his mother, but this was still not enough. Partly to blame is his system which can take and analyze pictures of cucumbers that are only 80×80 pixels in resolution.

“When I did a validation with the test images, the recognition accuracy exceeded 95%. But if you apply the system with real use cases, the accuracy drops down to about 70%. I suspect the neural network model has the issue of “overfitting” (the phenomenon in a neural network where the model is trained to fit only to the small training dataset) because of the insufficient number of training images.”

Makoto needs to invest a lot more than $1,000 — the money he spent so far — to make his cucumber assembly line really stellar. Even so, his results are inspiring because they go to show just how much you can achieve if you set your mind to it.



Science finds the most and least addictive foods

Scientists from the University of Michigan have found which are the most and least addictive foods in the world. They gathered data from over 500 participants and found that the most addictive foods are (no surprise) pizza, ice cream and chocolate, while the least addictive ones are cucumbers, carrots, beans and rice.

Pizza was the most addictive food, according to questionnaires answered by 400 people.


It’s been debated for years whether or not food addiction actually exists; naturally, we are all addicted to food in the sense that we have to eat in order to survive. But can you actually be addicted to certain foods, like hamburgers? There is still no general consensus on this, but biologists seem to dismiss this idea, while many psychologists claim that food addiction is a real, serious problem – there are documented cases with people going through withdrawal-like symptoms when living without certain foods. With this in mind, a researcher from the University of Michigan and one from the New York Obesity Research Center, the Department of Medicine set out to find what are the most addictive foods.

For this, they asked participants to answer questions based on the Yale Food Addiction Scale. The scale was designed in 2009 by Yale University’s Rudd Center for Food Policy and Obesity and asks people to answer 25 questions on how much they like a certain food. The scale asks participants to count the number of times they’ve agreed with sentences like, “I eat to the point where I feel physically ill” or “I spend a lot of time feeling sluggish or fatigued from overeating,” to help them identify the biggest offenders. Scientists emphasized that “foods” doesn’t mean only unprocessed foods like fruits and vegetables, but can also apply to processed foods.

However, when the same study was conducted on undergrads, chocolate turned out to be the most addictive food.

Study 1 – the undergraduates

They conducted two separate studies to see what foods are considered problematic – how much is a certain food overeaten or eaten up to the point where it causes physical discomfort. The first study was conducted on 120 undergraduates. who were recruited from flyers on campus or through the University of Michigan Introductory Psychology Subject Pool. Students received either financial compensation or study credit for their time.

No surprises there, chocolate took the top spot, with over 1 in 4 people considering chocolate problematic. Ice cream, french fries and pizza followed, again, rather expectedly. But there were also some surprises: breakfast cereals were more problematic than soda or fried chicken, while water was considered to be more problematic than cucumbers or beans… I guess no one really loves beans.


“As hypothesized, highly processed foods (with added fat and/or refined carbohydrates) appeared to be most associated with behavioral indicators of addictive-like eating,” the study writes.

Study 2

Ice is always one of the favorites.

The team also conducted a second study, on almost 400 participants.

“A total of 398 participants were recruited using Amazon’s Mechanical Turk (MTurk) worker pool to complete a study about eating behaviors and were compensatedfor their time”.


So this is the chart of what can be considered the most addictive foods. Interestingly enough, results were slightly different. Pizza took the top spot and chocolate had to settle for second. Chips, cookies and ice cream come closely after. Breakfast cereal dropped significantly, and the least popular food is… the cucumber.

I was surprised to find bananas close to the bottom of the list, even under water. But what’s really the takeaway here is that virtually all the addictive foods are processed.

“In summary, the current study found that highly processed foods, with added amounts of fat and/or refined carbohydrates (e.g., sugar, white flour), were most likely to be associated with behavioral indicators of addictive-like eating. Additionally, foods with high GL were especially related to addictive-like eating problems for individuals endorsing elevated symptoms of “food addiction.” Individuals endorsing symptoms of addictive-like eating behavior may be more susceptible to the large blood sugar spike of high GL foods, which is consistent with the importance of dose and rate of absorption in the addictive potential of drugs of abuse,” the study concludes.