Developers' Corner

The parameter "Greediness" defines a tradeoff between speed and completeness. On the one hand, the search speed can be increased by discarding candidates earlier based on a less strict criterion, while on the other hand, it might occur that no…

Classification is a very important method, often used in image processing. It can be utilized for directly segmenting an image into known classes, i.e., pixels are classified and grouped into regions, which can then be processed further. Important…

Everyone knows the issue, when you have numeric tuples with many elements, and you are interested in a quick overview of their content. Since HALCON 12, a new feature has been available to visualize numeric tuples graphically. In the Variable View,…

One of the most powerful tools in HDevelop is the Feature Histogram. The Feature Histogram is a very useful tool to get a feeling for the features of regions or contours. In many applications, it is necessary to select regions or contours depending…

A recommended practice for setting up classification tasks and OCR: Regularly review your growing set of training data with the HDevelop OCR Training File Browser – while developing the segmentation part, but also at further development steps. It is…

In the Developers' Corner article "Using regularization in MLP classification", we already described the use of regularization during the training of a multilayer perceptron (MLP) to enhance its capabilities of generalization to unknown data. In some…

When training any classifier, the goal is to train it so that it generalizes well for unknown data. However, in a real application, overfitting is often a problem, i.e., the classifier adapts too well to the training data and is not able to classify…