Regarding douobts about for storing quality of an image

[pandian]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) - jetson anno
• DeepStream Version - 6.01
• JetPack Version (valid for Jetson only) - 4.6.3
• TensorRT Version - 8.4
• NVIDIA GPU Driver Version (valid for GPU only) - 10.2
• Issue Type( questions, new requirements, bugs) - now we are implementing face recognition using ultra light weight model 1 mb so we have to store the face image for analytics purpose and we are storing an images but it is not that much of quality of an image so we need to store perfect iface mages for doing some analytics what can we do for storing quality of an image and if you give some technique about this it will be more helpful for us to doing analytics.
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

[yuweiw]

Could you describe your pipeline and how to save image now?

[pandian]

n_frame = pyds.get_nvds_buf_surface(hash(gst_buffer), frame_meta.batch_id)
frame = np.array(n_frame, copy=True, order=‘C’)
bgr_frame = cv2.cvtColor(frame, cv2.COLOR_RGBA2BGR)

            if obj_meta.class_id == 1:


                gray_frame = cv2.cvtColor(frame, cv2.COLOR_RGBA2GRAY)


                left = int(obj_meta.rect_params.left)
                top = int(obj_meta.rect_params.top)
                width = int(obj_meta.rect_params.width)
                height = int(obj_meta.rect_params.height)
                
                rect = dlib.rectangle(left, top, left + width, top + height)
                # Align the face image
                face_aligned = face_aligner.align(frame, gray_frame, rect)

                margin = 8
                roi = bgr_frame[max(rect.top() - margin,0):min(rect.bottom()+margin,bgr_frame.shape[0]),max(rect.left()-margin,0):min(rect.right()+margin,bgr_frame.shape[1])]

                resize_img = cv2.resize(roi,(500,500))
                cv2.imwrite("face_image/{}.jpg".format(count),resize_img)

[yuweiw]

You can try comparing different resize algorithms by setting the interpolation parameter for cv2.resize. But it’s hard to improve the quality of the image because the origin roi is too small.

[pandian]

do you know any other method for storing quality of an image

[yuweiw]

If the original image is relatively small, it is difficult to resize it too large and maintain the clarity.

[pandian]

ok tq