• 5 megapixel camera
• Color scanning
• Up to 6,400 steps per rotation plate
• Resolution 0.25 mm
• Volume scannable 200x200x150 mm
• assisted calibration
• Supply voltage: 12 VDC / 700 mA

Compared with the single camera technology, that at the camera LASER has the following advantages:

- Higher contrast image because the camera shots are obtained by the reflection of the laser light, so the resulting image is good even in low light; allows a greater detail, due to the fact that it does not analyze a simple frame, but a very bright line on the object surface, therefore allows to also detect small irregularities of the surface and provide a better picture of the variations in shape;

- It manages to get the scanning also of objects by shape and smooth surface;

- It does not require the application of a background to the shooting scene;

- It requires less computing power for the construction of the three-dimensional model.

To capture the object's shape, our system performs a superficial analysis of the same, causing it to rotate on a plate or fraction of a degree at a time while he is pointed at a linear laser light; The laser projects vertical lines covering the entire height of the object to be scanned and the reflected light reaches a camera that captures the resulting images and passes the appropriate software. To acquire all the necessary information, the object is rotated by an angle at the discretion, although normally it makes a full rotation (360 degrees). The program allows you to determine how many steps must carry the motor NEMA 17 for a full turn (360 °) until 6400; our stepper-motor is 200 steps / rev and piloted with the shield, makes a step every 32 pulses on STEP joints from Raspberry Pi 2; then step 200 multiplied by 32, are 6.400. The linear beams of laser are mutually inclined, in the horizontal plane, of 60 degrees, then the camera, being centered with respect to the two LASER, is at an angle of 30 ° with respect to the beam of each. The scanner must operate in a not excessively lit environment, so that there is a clear contrast between the area affected by the line of each laser and the object surface to be scanned. When you start scanning, the Raspberry Pi 2 starts recording images (frames) provided by the camera, with a periodicity set by the software: the default is 800 per full turn of the plate and then the object, but we have tried to work even with 2,000 frames / rev.

The need to adopt a second LASER arises when you have to scan for example a cube: in this case, in fact, only using a LASER are created some shadow areas due to the misalignment (ie of the different angle with which the two elements turn the piece to be scanned) between the camera and the laser. In the specific case, since the two on the same axis, during rotation there are areas in which the beam is obscured by one of the walls and does not reach the camera, preventing to perceive the reflected beam and therefore causing "holes" in the acquired model . If we add a second LASER phase shifted by 30 degrees in the direction of counterclockwise rotation, the shadow areas disappear: in fact, when the object rotates exceeding the critical angle of the first laser, the shadow area is scanned through the second LASER. For the acquisition of objects that do not have these facets to hide the laser line to the eye of the camera, you can use it to scan, only a LASER. Let's see how the acquisition takes place: in each frame tripped, is considered to be the reflection of the laser line, which is vertical and is a thin slice of the surface; at each rotation you are then obtained one frame with an illuminated slice falling in an always different position, up to complete the desired surface.

The Raspberry Pi 2 analyzes, for each frame, the lit area, discarding the rest of the image, it needs just to check the correct sequence of queuing of slices; putting together the slices you rebuild the image.

We now move to the mechanical structure of the scanner, which is the set of plastic molded parts (for example with our 3Drag) assembled by means of threaded rods in iron and bolts; the main blocks are the base, which contains in the center the electronics and the camera and to the sides the two linear LASER, and the plate that supports the object, rotatable on a bush bound to a plastic support connected by means of the threaded bars to basis. The plate can be obtained by a plexiglass disc or forex 4 mm thick, but also of PLA and then printed; after having applied a black sheet covering, individuatene the center and return it in the form of a point with a permanent marker. The linear LASER are to red light (650 nm) wavelength of Class 3A (cod. 2510-LASER5MWLINEA); each develops an optical power of 3 mW and requires a power supply of DC voltage of 4 ÷ 6 V (absorbs 280 mA). The LASER integrates a collimating aspheric lens to direct the light at a single point, which is then transformed in line thanks to an acrylic cylindrical lens placed at the exit of the laser beam, so as to create a light beam with a triangular section with a ' opening of about 45 °

The software implements a web interface, through which we can work with your scanner, or make settings and acquisitions. For each capture file is generated.

FreeLSS implements:

• Real-time preview of the scan;
• assisted calibration;
• support for the two-mode laser;
• up to 6,400 steps per rotation of the plate;
• Camera Support Still mode and Video mode;
• setting of image processing parameters;
• possibility to obtain partial images to debug if scanning fails;
• Support for manual movement of the turntable.