Generating graphical representation of scanned objects

A system comprising:

• at least one processor;
• and
• a non-transitory computer-readable storage medium comprising instructions that are executable by the at least one processor to cause the system to:
• receive a digital image that represents an object scanned by a detection device of a given security system;
• determine that the digital image comprises a first item of a first classification of interest utilizing at least one neural network, that is configured to perform object detection on the digital image to localize a target within the digital image followed by performing a classification task based on the localized target;
• determine that the digital image comprises a second item of a second classification of interest that differs from the first classification of interest utilizing a classification technique that differs from utilizing the at least one neural network;
• based on determining that (i) the first item of the first classification of interest and (ii) the second item of the second classification of interest are deemed to be a security threat for the given security system, generate, for output at a display, a visualization for the object scanned by the detection device that comprises (i) at least a portion of the digital image, (ii) a first indicator indicating a location of the first item within the digital image, wherein the first indicator comprises a first visualization characteristic corresponding to the first classification of interest, and (iii) a second indicator indicating a location of the second item within the digital image, wherein the second indicator comprises a second visualization characteristic corresponding to the second classification of interest, wherein the first visualization characteristic and the second portion visualization characteristic differ;
• and
• cause the display to output the visualization for the object scanned by the detection device.

A non-transitory computer-readable storage medium comprising instructions that are executable by at least one processor to cause a computing system to:

• receive a digital image that represents an object scanned by a detection device of a given security system;
• determine that the digital image comprises a first item of a first classification of interest utilizing at least one neural network that is configured to perform object detection on the digital image to localize a target within the digital image followed by performing a classification task based on the localized target;
• determine that the digital image comprises a second item of a second classification of interest that differs from the first classification of interest utilizing a classification technique that differs from utilizing the at least one neural network;
• based on determining that (i) the first item of the first classification of interest and (ii) the second item of the second classification of interest are deemed to be a security threat for the given security system, generate, for output at a display, a visualization for the object scanned by the detection device that comprises (i) at least a portion of the digital image, (ii) a first indicator indicating a location of the first item within the digital image, wherein the first indicator comprises a first visualization characteristic corresponding to the first classification of interest, and (iii) a second indicator indicating a location of the second item within the digital image, wherein the second indicator comprises a second visualization characteristic corresponding to the second classification of interest, wherein the first visualization characteristic and the second visualization characteristic differ;
• and
• cause the display to output the visualization for the object scanned by the detection device.