public static ParseObject TFSRequestRestful(Bitmap bImage)
{
ParseObject pObj = null;
int im_width = bImage.Width;
int im_height = bImage.Height;
var httpWebRequest =
(HttpWebRequest)WebRequest.Create(ConfigurationManager.AppSettings["ServerHost2"]);
httpWebRequest.ContentType = "application/json";
httpWebRequest.Method = "POST";
using (var streamWriter = new StreamWriter(httpWebRequest.GetRequestStream()))
{
/*NDArray bitmapBytes = GetBitmapBytes2NDArray(bImage);*/
string bitmapBytes = GetBitmapBytes2String(bImage);
string json = "{\"instances\": " + bitmapBytes + "}";
streamWriter.Write(json);
streamWriter.Flush();
streamWriter.Close();
}
var httpResponse = (HttpWebResponse)httpWebRequest.GetResponse();
using (var streamReader = new StreamReader(httpResponse.GetResponseStream()))
{
var result = streamReader.ReadToEnd();
//Console.WriteLine(result.ToString());
JObject jobj = JObject.Parse(result);
var dtNumDetect = jobj["predictions"][0]["num_detections"].ToObject <int>();
var dtClassList = jobj["predictions"][0]["detection_classes"].Select(x => (int)x).ToArray();
var dtScoreList = jobj["predictions"][0]["detection_scores"].Select(x => (float)x).ToArray();
// if only obj is human, it will send data
float threshold = float.Parse(ConfigurationManager.AppSettings["Threshold"]);
pObj = new ParseObject();
//int ppc = 0;
for (int pid = 0; pid < dtNumDetect; pid++)
{
if ((dtClassList[pid] == 1) && (threshold <= dtScoreList[pid]))
{
pObj.NumOfDetect += 1;
//pObj.ScoreList.SetValue(dtScoreList[pid], ppc);
Boxes boxes = new Boxes();
var dtBoxesList = jobj["predictions"][0]["detection_boxes"][pid].Select(x => (float)x).ToArray();
boxes.BoxX1 = dtBoxesList[0] * im_height; //p1_height
boxes.BoxX2 = dtBoxesList[1] * im_width; //p1_width
boxes.BoxY1 = dtBoxesList[2] * im_height; //p2_height
boxes.BoxY2 = dtBoxesList[3] * im_width; //p2_width
pObj.BoxList.Add(boxes);
//ppc++;
}
}
}
return(pObj);
}
public static ParseObject TestTFSRequestRestful(Bitmap bImage)
{
ParseObject pObj = null;
int im_width = bImage.Width;
int im_height = bImage.Height;
pObj = new ParseObject();
string res = "{\"predictions\": [{\"detection_scores\": [0.992537796, 0.0], " +
"\"detection_classes\": [1.0, 1.0], " +
"\"num_detections\": 1.0, " +
"\"detection_boxes\": [[0.0457661822, 0.0729743466, 0.957826436, 0.988076806], [0.0, 0.0, 0.0, 0.0]]}]}";
JObject jobj = JObject.Parse(res);
var dtNumDetect = jobj["predictions"][0]["num_detections"].ToObject <int>();
var dtClassList = jobj["predictions"][0]["detection_classes"].Select(x => (int)x).ToArray();
var dtScoreList = jobj["predictions"][0]["detection_scores"].Select(x => (float)x).ToArray();
float threshold = float.Parse(ConfigurationManager.AppSettings["Threshold"]);
for (int pid = 0; pid < dtNumDetect; pid++)
{
if ((dtClassList[pid] == 1) && (threshold <= dtScoreList[pid]))
{
pObj.NumOfDetect += 1;
//pObj.ScoreList.SetValue(dtScoreList[pid], ppc);
Boxes boxes = new Boxes();
var dtBoxesList = jobj["predictions"][0]["detection_boxes"][pid].Select(x => (float)x).ToArray();
boxes.BoxX1 = dtBoxesList[0] * im_height; //p1_height
boxes.BoxX2 = dtBoxesList[1] * im_width; //p1_width
boxes.BoxY1 = dtBoxesList[2] * im_height; //p2_height
boxes.BoxY2 = dtBoxesList[3] * im_width; //p2_width
pObj.BoxList.Add(boxes);
//ppc++;
}
}
return(pObj);
}