Humans have an innate knack for identifying food that is fit to eat. There’s a reason you instinctively enjoy fresh fruit and vegetables, but find maggot-infested rotting flesh offputting, for example. However, we like to automate as much of the food production process as possible so we can do other things, so it’s necessary to have machines sort the ripe and ready produce from the rest at times. [kutluhan_aktar] has found a way to do just that, using the power of neural networks.
The project’s goal is a straightforward one, aiming to detect ripeness in fruit and vegetables by monitoring pigment changes. Rather than use a camera, the project relies on data from an AS7341 visible light sensor, which is better suited to capturing accurate spectral data. This allows a better read of the actual light reflected by the fruit, as determined by the pigments in the skin which are directly related to ripeness.
Sample readings were taken from a series of fruit and vegetables over a period of several days, which allowed a database to be built up of the produce at various stages of ripeness. This was then used to create a TensorFlow model which can determine the ripeness of fruit held under the sensor with a reasonable degree of certainty.
The build is a great example of the use of advanced sensing in combination with neural networks. We suspect the results are far more accurate than could have reasonably be determined with a cheap webcam, though we’d love to see an in-depth comparison as such.
Believe it or not, it’s not the only fruit spectrometer we’ve featured in these hallowed pages. Video after the break.
Continue reading “Detecting Ripeness In Fruit And Vegetables Via Neural Networks” →
Our bodies are not like LEGO blocks or computers because we cannot swap out our parts in the living room while watching television. Organ transplants and cosmetic surgery are currently our options for upgrades, repairs, and augments, but post-transplant therapy can be a lifelong commitment because of rejection. Elective surgery costs more than a NIB Millenium Falcon LEGO set. Laboratories have been improving the processes and associated treatments for decades but experimental labs and even home laboratories are getting in on the action as some creative minds take the stage. These folks aren’t performing surgeries, but they are expanding what is possible to for people to do and learn without a medical license.
One promising gateway to human building blocks is the decellularization and recellularization of organic material. Commercial scaffolds exist but they are expensive, so the average tinkerer isn’t going to be buying a few to play with over a holiday weekend.
Let’s explore what all this means. When something is decellularized, it means that the cells are removed, but the structure holding the cells in place remains. Recellularizing is the process where new cells are grown in that area. Decellularizing is like stripping a Hilton hotel down to the girders. The remaining structures are the ECM or the Extra Cellular Matrix, usually referred to as scaffolding. The structure has a shape but no functionality, like a stripped hotel. The scaffolding can be repopulated with new cells in the same way that our gutted hotel can be rebuilt as a factory, office building, or a hospital.
Continue reading “Decellularization: Apples To Earlobes” →
We thought we were going to read an article about, perhaps, a quadcopter that could fetch beer, or donuts. What we got was more along the lines of a donut dragging itself across the floor, rendering it pitiful and advisibly indigestible.
Sometimes people joke about not wanting to get in mind of a crazy person. We understand. While we could certainly follow [Michael Kohn]’s logic, the motivation was alien. Either way, in a rare turn of events there was not a single Arduino to be seen; just reverse engineering, unique solutions, and even a custom board. This is what some of you have been asking for… we think.
The brain of the questionable contraption is a TI MSP430G2231 and a tiny forward only motor driver circuit. The MSP waits for a signal from a hacked IR remote control from a cheap RC car. It then turns those into the appropriate motor control signals which go to some of those nice tiny metal gearboxes.
There were, naturally, a lot of technical issues in mounting the electronics to the food that, well… they didn’t need to be solved, but they were solved. For example, masking tape apparently does not stick well to green peppers, so toothpicks must be employed to pin the tape in place. Hopefully knowledge like this is scheduled for the nightly wipe while we sleep, but we’ll probably hold onto it till we die, unlike expensive piano lessons.
In the end we had a good laugh, and the idea is so dumb it will probably be an educational Kickstarter next week. Video after the break.
Continue reading “Technically A Hack. Still Questionable. Remote Control Food.” →
It’s great having fresh vegetables just a few steps away from the kitchen, but it takes work to keep those plants healthy. [Pierre] found this out the hard way after returning from vacation to find his tomato plant withering away. He decided to put an end to this problem by building his own solar-powered plant watering system (page in French, Google translation).
An Arduino serves as the brain of the system. It’s programmed to check a photo resistor every ten minutes. At 8:30PM, the Arduino will decide how much to water the plants based on the amount of sunlight it detected throughout the day. This allows the system to water the plants just the right amount. The watering is performed by triggering a 5V relay, which switches on a swimming pool pump.
[Pierre] obviously wanted a “green” green house, so he is powering the system using sunlight. A 55 watt solar panel recharges a 12V lead acid battery. The power from the battery is stepped down to the appropriate 5V required for the Arduino. Now [Pierre] can power his watering system from the very same energy source that his plants use to grow.
[Eric Maundu] is farming in Oakland. There are no open fields in this concrete jungle, and even if there were the soil in his part of town is contaminated and not a suitable place in which to grow food. But he’s not using farming methods of old. In fact farmers of a century ago wouldn’t recognize anything he’s doing. His technique uses fish, circulated water, and gravel to grow vegetables in whatever space he can find; a farming method called aquaponics.
The video after the break gives an excellent look at his farm. The two main parts of the system are a large water trough where fish live, and a raised bed of gravel where the fish waste in the water is filtered out and composted by bacteria to becomes food for the vegetables. More parts can be added into the mix. For instance, once the water has been filtered by the stone bed it can be gravity fed into another vessel which is being used to grow lettuce suspended by floating foam board. But the water always ends up back in the fish trough where it can be reused. This ends up saving anywhere from 90-98% of the water used in normal farming.
But [Eric] is also interested in adding some automation. About seven minutes into the video we get a look at the control systems he’s working on with the help of Arduino and other hardware.
Continue reading “Urban Farming Uses Aquaponics To Make Farmland Where There Is None” →
The Cheap Vegetable Gardner wanted more automation than their previous PS2 controller based grow system. This time they set out to design a full featured, compact grow controller that can measure temperature and humidity as well as control a heat lamp, fan, and water pump. An Arduino provides USB connectivity and interfaces the solid state relays and sensors. The assembled project all fits in a box but we are left wondering how much heat the four SSRs generate and will it be a problem?