Unknown Encoding for Data Matrix encoded with google Barcode Scanner (Vision)

Question

I'm trying to scan some Data Matrix bar codes with my implementation of the Vision Barcode Scanner, but I'm having issues on the encoded ones.

I'm using:

int codesDetected = Barcode.ALL_FORMATS;

But it still returns Unknown encoding for those ones.

On the other hand, Google Goggles app scan them properly, same for some other bar code readers using zxing.

Example of bar codes:

I'm willing to move to Zxing if that solves the issue, but I want to know if there is something I'm missing in my implementation (I took the Vision sample app) for these type of bar codes (data matrix encoded).

Thanks

Beret

I get into this google sample and this returns "Unknown encoding" as well. I think that the barcodes are corrupted at some point. Zixing application from play google gives wired result. For small one this shows that text "123456789ðñòóôõö÷øùúûüýþÿ". – Acrogen

I think Zxing will not solve that problem. What I found that Zxing is giving wrong out put for such scenario. Same result for Google Goggles. (I hove you already tested these scenario) Google Vision sample app also have problem for your image and other image also. But for other data matrix images I observed google vision sometimes able to decode properly (giving correct value). – Alexipharmic

Scanning those codes with a Honeywell Voager 1450g imager detects the code type, but the data in it seem to be corrupt, e.g. the small code returns this: w123456789???????????????? (the leading w indicates the barcode type). – Longo

For completeness sake, the large code decodes as w123456789????????????????123456789????????????????123456789????????????????123456789????????????????123456789????????????????123456789????????????????123456789????????????????123456789????????????????123456789????????????????123456789???????????????? – Longo

The smaller code contains bytes 0x31 ... 0x39 0xf0 .. 0xff in total 9 + 16 = 25 bytes. The content represented by characters of course depends on the encoding. It's not valid UTF8 / Unicode. Represented by Windows1252 (Western Europe) it's 123456789ðñòóôõö÷øùúûüýþÿ. The large code contains the same repeated 10 times. The codes are not "corrupted". Both are valid Datamatrix containing binary data. – Gravelblind