• What hardware is required?
  A basic recognition system needs just a standard PC (minimum 800MHz processor), a camera and a video acquisition board if you wish to use direct input images, and the Spikenet software (SNVision Library). In the case of more demanding problems with large images and high number of targets, you can implement Spikenet on parallel hardware.

• Does SNVision SDK require a hardware key?
  You will need a hardware key to use the SDK which includes SNVision Model Builder and SNVision library. We can provide you with keys for paralell ports or USB connectors.
  
  

• What is the difference between SNVision Demo and SNVision Model Builder?
  SNVision Demo is a free trial version of SNVision Model Builder. It allows you to open images, learn some models and test the system performance with noise, blur, luminance, contrast changes, rotations and zooms. It does not allow you to display or save recognition results nor save and reload model files. To do so, you will need the full version: SNVision Model Builder.

• What image formats are supported?
  More than 60 raster formats are supported including: bmp, jpg, pcx, tif, etc...You can also use video sequences saved as AVI files.

• Why am I unable to open a GIF file with SNVision Model Builder?
  The GIF patent has expired in the US. It is still valid in other countries, yet not for very long. Access to the GIF format is restricted to people who have a license from Unisys. To open your image file with SNVision MB, save it under another format, for example as a jpeg file.

• Can I use a video source as an input?
  You can use a direct video input. For example, you can use a webcam connected to a USB port, or a firewire compatible digital camera or webcam connected to a IEEE 1934 port.
  This feature can be activated via the "video" button for SN Demo and the "Capture" option for SNVisionMB (File Menu). It acquires images from the current Windows default input.

• The recognition mechanism does not seem to tolerate changes in orientation and zoom that are higher than 4%. What can I do ?
  SpikeNet's recognition mechanism can tolerate variations in rotation and size of a few percents with one single prototype for each target. If you wish to increase this level of tolerance, you can automatically generate multiple models to cover a wider range of zoom factors and/or orientations. When you open the Learn Window, by default the Zoom and Rotate factors boxes are left unchecked. If you check them, SNVision MB will generate multiple models to cover the range of zoom factors and/or orientations that you specify.

• What is propagation?
  The propagation is a recognition parameter. It is the amount of information extracted from the input image. When you increase the propagation value, more information about the image is provided thus improving the signal to noise ratio of the recognition process. But as more information is provided, the processing time goes up and you increase the number of false positives. The propagation value should be kept below 30.
  

• What is Contrast Minimum?
  SpikeNet’s algorithm is designed to be extremely good at detecting objects even when the contrast in the image is very low. One problem that this can cause is that in an image where the only interesting objects are at high contrasts, unwanted detections may occur in regions of very low contrast. To avoid detecting details which are not relevant, you can raise the Minimum Contrast value that is normally fixed at 0. This rules out the false detections occurrences in regions with very low contrast and greatly speeds up the processing time. This way, only the regions in which the contrast is above the Minimum Contrast value are subject to processing.
  

• What is the threshold?
  The threshold is a model parameter which adjusts the sensitivity of the model. Increasing the threshold will increase the sensitivity of the model, generating hits that better match your model. The higher the threshold value is, the more similar to the model an object has to be in order to be detected
  Lowering the threshold will lower the sensitivity increasing the amount of error. This can be useful when you have used a single model (a face for example) and want the system to respond even when the view is not identical to that of the model used.
  

• What is Detail?
  It is the amount of information contained in a model. Increasing the detail value will improve the signal to noise ratio, but with the penalty of increasing the processing time.
  

• Does the Model size affect processing time?
  No, the size of the model has no impact on the processing time. But, the amount of detail that you specify does.
  

• What is the best size for the models?
  To increase the robustness it is best to create models of 30x30 pixels. To do so you will probably need to lower the resolution of your source image using the Scale parameter.
  

• What is the Throughput value shown in the Processing Information Window?
  The throughput value refers to number of Megapixels processed per second.
  Here is how it is determined: (number of pixels processed within the image) x (number of models)/processing time in seconds.
  

• Why is the Frame Rate value not equal to the Processing Time value?
  The Frame rate refers to the actual number of images processed by SNVision MB. It includes the time to acquire the data and display the image.
  
  
• Which memory size is required to integrate SNVision Library in an application?
  The memory size required depends on many parameters but mainly on the input image size and the models number.
  The formula determining the memory size is:
  1500 Kb + input image pixels * 2/1024 Kb + models number * 5Kb
  
  

• What do I do to try?
  You can download SNvision Demo for a first quick test of SpikeNet technology. SNVision Demo is trial version of SNVision Model Builder, an application that enables you to create the models of the targets you wish to locate and to test the recognition process. If you would like to start developing an application using SpikeNet Technology, contact us so that we can arrange to send you our best proposal for a time limited trial version of SNVision SDK. Please provide information about your vision application and your hardware requirements. SNVision SDK requires a hardware key.

• Which products will I need to develop a vision application using SpikeNet Technology?
  All you need is SNVision SDK which includes SNVision Model Builder to create the models you want to locate and to test the robustness of the recognition process, and SNVision Library (dll).
  
  Once you have developed your vision application, if you want to distribute SNVision Libary, you will need a licensing agreement. Contact us at sales@spikenet-technology.com to obtain a proposal.

You didn't find an answer to your question? Send us an email at:

support@spikenet-technology.com