应用第三方ByteTrack实现目标跟踪

在上一篇博文，我们实现了应用官网ByteTrack实现的目标跟踪。但吹毛求疵地说，官网的ByteTrack有一些不足：1、为了通用性及科研的要求，代码过于冗长，但这不利于集成到自己的程序中；2、目标跟踪结果没有目标类别的信息，需要自己额外添加；3、再次运行算法时，ID不会复位，它会接着上次的ID继续排序。因此，我们完全有理由对其进行优化。

在github中，有很多优化后的ByteTrack，所以我们没有必要自己去优化。在这里我们选择下面这个ByteTrack：

https://github.com/jahongir7174/ByteTrack/

为此，我们把核心函数整合到一个名为track.py的文件内，并把其放入到mytrack的文件夹内，然后把该文件夹复制到当前目录下。我把track.py文件内的完整内容复制到本博文的后面。

为使用该文件，先要导入：

from mytrack import track

然后实例化ByteTrack：

bytetracker = track.BYTETracker(fps)

BYTETracker只需要一个输入参数，其表示视频的帧率，默认为30。

使用BYTETracker也很简单：

tracks = bytetracker.update(boxes, scores, object_classes)

三个输入参数都为目标检测得到的每一帧的检测结果，数据类型为numpy.array。boxes为目标边框的左上角和右下角坐标，scores为目标置信值，object_classes为目标类型。

输出参数为该帧的跟踪结果，前4个元素为目标边框的左上角和右下角坐标（其与boxes有所不同），第4个元素为跟踪ID，第5个元素为目标得分值，第6个元素为目标类型，第7个元素为boxes所对应的索引值。

可以看出该第三方ByteTrack与官网的ByteTrack使用方法区别不大。

为使用该ByteTrack，只需要安装以下软件包即可：

conda install python=3.8
pip install scipy -i https://pypi.tuna.tsinghua.edu.cn/simple
conda install -c conda-forge lap

从中可以看出，该第三方ByteTrack所需的软件包很少，安装起来也不会出现各类莫名其妙的问题。

下面就可以编写目标跟踪代码了。在这里，我们只把if outputs is not None:内的部分展示出来，而其他部分与上一篇博文的代码是一样的：

for output in outputs:x1, y1, x2, y2 = list(map(int, output[:4]))boxes.append([x1, y1, x2, y2])confidences.append(output[4])object_classes.append(output[5])tracks = bytetracker.update(np.array(boxes), np.array(confidences), np.array(object_classes))if len(tracks) > 0:identities = tracks[:, 4]object_classes = tracks[:, 6]idxs = tracks[:, 7]for i, identity in enumerate(identities):if object_classes[i] == 2:box_label(frame, outputs[int(idxs[i]),:4], '#'+str(int(identity))+' car' , (167, 146, 11))elif object_classes[i] == 5:box_label(frame, outputs[int(idxs[i]),:4], '#'+str(int(identity))+' bus', (186, 55, 2))elif object_classes[i] == 7:box_label(frame, outputs[int(idxs[i]),:4], '#'+str(int(identity))+' truck', (19, 222, 24))

下面是完整的track.py内容：

import numpy
import lap
import scipydef linear_assignment(cost_matrix, thresh):# Linear assignment implementations with scipy and lap.lapjvif cost_matrix.size == 0:matches = numpy.empty((0, 2), dtype=int)unmatched_a = tuple(range(cost_matrix.shape[0]))unmatched_b = tuple(range(cost_matrix.shape[1]))return matches, unmatched_a, unmatched_b_, x, y = lap.lapjv(cost_matrix, extend_cost=True, cost_limit=thresh)matches = [[ix, mx] for ix, mx in enumerate(x) if mx >= 0]unmatched_a = numpy.where(x < 0)[0]unmatched_b = numpy.where(y < 0)[0]return matches, unmatched_a, unmatched_bdef compute_iou(a_boxes, b_boxes):"""Compute cost based on IoU:type a_boxes: list[tlbr] | np.ndarray:type b_boxes: list[tlbr] | np.ndarray:rtype iou | np.ndarray"""iou = numpy.zeros((len(a_boxes), len(b_boxes)), dtype=numpy.float32)if iou.size == 0:return ioua_boxes = numpy.ascontiguousarray(a_boxes, dtype=numpy.float32)b_boxes = numpy.ascontiguousarray(b_boxes, dtype=numpy.float32)# Get the coordinates of bounding boxesb1_x1, b1_y1, b1_x2, b1_y2 = a_boxes.Tb2_x1, b2_y1, b2_x2, b2_y2 = b_boxes.T# Intersection areainter_area = (numpy.minimum(b1_x2[:, None], b2_x2) - numpy.maximum(b1_x1[:, None], b2_x1)).clip(0) * \(numpy.minimum(b1_y2[:, None], b2_y2) - numpy.maximum(b1_y1[:, None], b2_y1)).clip(0)# box2 areabox1_area = (b1_x2 - b1_x1) * (b1_y2 - b1_y1)box2_area = (b2_x2 - b2_x1) * (b2_y2 - b2_y1)return inter_area / (box2_area + box1_area[:, None] - inter_area + 1E-7)def iou_distance(a_tracks, b_tracks):"""Compute cost based on IoU:type a_tracks: list[STrack]:type b_tracks: list[STrack]:rtype cost_matrix np.ndarray"""if (len(a_tracks) > 0 and isinstance(a_tracks[0], numpy.ndarray)) \or (len(b_tracks) > 0 and isinstance(b_tracks[0], numpy.ndarray)):a_boxes = a_tracksb_boxes = b_trackselse:a_boxes = [track.tlbr for track in a_tracks]b_boxes = [track.tlbr for track in b_tracks]return 1 - compute_iou(a_boxes, b_boxes)  # cost matrixdef fuse_score(cost_matrix, detections):if cost_matrix.size == 0:return cost_matrixiou_sim = 1 - cost_matrixdet_scores = numpy.array([det.score for det in detections])det_scores = numpy.expand_dims(det_scores, axis=0).repeat(cost_matrix.shape[0], axis=0)fuse_sim = iou_sim * det_scoresreturn 1 - fuse_sim  # fuse_costclass KalmanFilterXYAH:"""A Kalman filter for tracking bounding boxes in image space.The 8-dimensional state spacex, y, a, h, vx, vy, va, vhcontains the bounding box center position (x, y), aspect ratio a, height h,and their respective velocities.Object motion follows a constant velocity model. The bounding box location(x, y, a, h) is taken as direct observation of the state space (linearobservation model)."""def __init__(self):ndim, dt = 4, 1.# Create Kalman filter model matrices.self._motion_mat = numpy.eye(2 * ndim, 2 * ndim)for i in range(ndim):self._motion_mat[i, ndim + i] = dtself._update_mat = numpy.eye(ndim, 2 * ndim)# Motion and observation uncertainty are chosen relative to the current# state estimate. These weights control the amount of uncertainty in# the model. This is a bit hacky.self._std_weight_position = 1. / 20self._std_weight_velocity = 1. / 160def initiate(self, measurement):"""Create track from unassociated measurement.Parameters----------measurement : ndarrayBounding box coordinates (x, y, a, h) with center position (x, y),aspect ratio a, and height h.Returns-------(ndarray, ndarray)Returns the mean vector (8 dimensional) and covariance matrix (8x8dimensional) of the new track. Unobserved velocities are initializedto 0 mean."""mean_pos = measurementmean_vel = numpy.zeros_like(mean_pos)mean = numpy.r_[mean_pos, mean_vel]std = [2 * self._std_weight_position * measurement[3],2 * self._std_weight_position * measurement[3],1e-2,2 * self._std_weight_position * measurement[3],10 * self._std_weight_velocity * measurement[3],10 * self._std_weight_velocity * measurement[3],1e-5,10 * self._std_weight_velocity * measurement[3]]covariance = numpy.diag(numpy.square(std))return mean, covariancedef predict(self, mean, covariance):"""Run Kalman filter prediction step.Parameters----------mean : ndarrayThe 8 dimensional mean vector of the object state at the previoustime step.covariance : ndarrayThe 8x8 dimensional covariance matrix of the object state at theprevious time step.Returns-------(ndarray, ndarray)Returns the mean vector and covariance matrix of the predictedstate. Unobserved velocities are initialized to 0 mean."""std_pos = [self._std_weight_position * mean[3],self._std_weight_position * mean[3],1e-2,self._std_weight_position * mean[3]]std_vel = [self._std_weight_velocity * mean[3],self._std_weight_velocity * mean[3],1e-5,self._std_weight_velocity * mean[3]]motion_cov = numpy.diag(numpy.square(numpy.r_[std_pos, std_vel]))# mean = np.dot(self._motion_mat, mean)mean = numpy.dot(mean, self._motion_mat.T)covariance = numpy.linalg.multi_dot((self._motion_mat, covariance, self._motion_mat.T)) + motion_covreturn mean, covariancedef project(self, mean, covariance):"""Project state distribution to measurement space.Parameters----------mean : ndarrayThe state's mean vector (8 dimensional array).covariance : ndarrayThe state's covariance matrix (8x8 dimensional).Returns-------(ndarray, ndarray)Returns the projected mean and covariance matrix of the given stateestimate."""std = [self._std_weight_position * mean[3],self._std_weight_position * mean[3],1e-1,self._std_weight_position * mean[3]]innovation_cov = numpy.diag(numpy.square(std))mean = numpy.dot(self._update_mat, mean)covariance = numpy.linalg.multi_dot((self._update_mat, covariance, self._update_mat.T))return mean, covariance + innovation_covdef multi_predict(self, mean, covariance):"""Run Kalman filter prediction step (Vectorized version).Parameters----------mean : ndarrayThe Nx8 dimensional mean matrix of the object states at the previoustime step.covariance : ndarrayThe Nx8x8 dimensional covariance matrix of the object states at theprevious time step.Returns-------(ndarray, ndarray)Returns the mean vector and covariance matrix of the predictedstate. Unobserved velocities are initialized to 0 mean."""std_pos = [self._std_weight_position * mean[:, 3],self._std_weight_position * mean[:, 3],1e-2 * numpy.ones_like(mean[:, 3]),self._std_weight_position * mean[:, 3]]std_vel = [self._std_weight_velocity * mean[:, 3],self._std_weight_velocity * mean[:, 3],1e-5 * numpy.ones_like(mean[:, 3]),self._std_weight_velocity * mean[:, 3]]sqr = numpy.square(numpy.r_[std_pos, std_vel]).Tmotion_cov = [numpy.diag(sqr[i]) for i in range(len(mean))]motion_cov = numpy.asarray(motion_cov)#print(mean)#print('eee')#print(self._motion_mat.T)mean = numpy.dot(mean, self._motion_mat.T)#print('fff')left = numpy.dot(self._motion_mat, covariance).transpose((1, 0, 2))covariance = numpy.dot(left, self._motion_mat.T) + motion_covreturn mean, covariancedef update(self, mean, covariance, measurement):"""Run Kalman filter correction step.Parameters----------mean : ndarrayThe predicted state's mean vector (8 dimensional).covariance : ndarrayThe state's covariance matrix (8x8 dimensional).measurement : ndarrayThe 4 dimensional measurement vector (x, y, a, h), where (x, y)is the center position, a the aspect ratio, and h the height of thebounding box.Returns-------(ndarray, ndarray)Returns the measurement-corrected state distribution."""projected_mean, projected_cov = self.project(mean, covariance)chol_factor, lower = scipy.linalg.cho_factor(projected_cov, lower=True, check_finite=False)kalman_gain = scipy.linalg.cho_solve((chol_factor, lower),numpy.dot(covariance, self._update_mat.T).T,check_finite=False).Tinnovation = measurement - projected_meannew_mean = mean + numpy.dot(innovation, kalman_gain.T)new_covariance = covariance - numpy.linalg.multi_dot((kalman_gain, projected_cov, kalman_gain.T))return new_mean, new_covariancedef gating_distance(self, mean, covariance, measurements, only_position=False, metric='maha'):"""Compute gating distance between state distribution and measurements.A suitable distance threshold can be obtained from `chi2inv95`. If`only_position` is False, the chi-square distribution has 4 degrees offreedom, otherwise 2.Parameters----------mean : ndarrayMean vector over the state distribution (8 dimensional).covariance : ndarrayCovariance of the state distribution (8x8 dimensional).measurements : ndarrayAn Nx4 dimensional matrix of N measurements, each informat (x, y, a, h) where (x, y) is the bounding box centerposition, a the aspect ratio, and h the height.only_position : Optional[bool]If True, distance computation is done with respect to the boundingbox center position only.metric : strDistance metric.Returns-------ndarrayReturns an array of length N, where the i-th element contains thesquared Mahalanobis distance between (mean, covariance) and`measurements[i]`."""mean, covariance = self.project(mean, covariance)if only_position:mean, covariance = mean[:2], covariance[:2, :2]measurements = measurements[:, :2]d = measurements - meanif metric == 'gaussian':return numpy.sum(d * d, axis=1)elif metric == 'maha':factor = numpy.linalg.cholesky(covariance)z = scipy.linalg.solve_triangular(factor, d.T, lower=True, check_finite=False, overwrite_b=True)return numpy.sum(z * z, axis=0)  # square mahaelse:raise ValueError('invalid distance metric')class State:New = 0Tracked = 1Lost = 2Removed = 3class Track:count = 0shared_kalman = KalmanFilterXYAH()def __init__(self, tlwh, score, cls):# wait activateself._tlwh = numpy.asarray(self.tlbr_to_tlwh(tlwh[:-1]), dtype=numpy.float32)self.kalman_filter = Noneself.mean, self.covariance = None, Noneself.is_activated = Falseself.score = scoreself.tracklet_len = 0self.cls = clsself.idx = tlwh[-1]def predict(self):mean_state = self.mean.copy()if self.state != State.Tracked:mean_state[7] = 0self.mean, self.covariance = self.kalman_filter.predict(mean_state, self.covariance)@staticmethoddef multi_predict(tracks):if len(tracks) <= 0:returnmulti_mean = numpy.asarray([st.mean.copy() for st in tracks])multi_covariance = numpy.asarray([st.covariance for st in tracks])for i, st in enumerate(tracks):if st.state != State.Tracked:multi_mean[i][7] = 0multi_mean, multi_covariance = Track.shared_kalman.multi_predict(multi_mean, multi_covariance)#print('eee')for i, (mean, cov) in enumerate(zip(multi_mean, multi_covariance)):tracks[i].mean = meantracks[i].covariance = covdef activate(self, kalman_filter, frame_id):"""Start a new tracklet"""self.kalman_filter = kalman_filterself.track_id = self.next_id()self.mean, self.covariance = self.kalman_filter.initiate(self.convert_coords(self._tlwh))self.tracklet_len = 0self.state = State.Trackedif frame_id == 1:self.is_activated = Trueself.frame_id = frame_idself.start_frame = frame_iddef re_activate(self, new_track, frame_id, new_id=False):self.mean, self.covariance = self.kalman_filter.update(self.mean, self.covariance,self.convert_coords(new_track.tlwh))self.tracklet_len = 0self.state = State.Trackedself.is_activated = Trueself.frame_id = frame_idif new_id:self.track_id = self.next_id()self.score = new_track.scoreself.cls = new_track.clsself.idx = new_track.idxdef update(self, new_track, frame_id):"""Update a matched track:type new_track: Track:type frame_id: int:return:"""self.frame_id = frame_idself.tracklet_len += 1new_tlwh = new_track.tlwhself.mean, self.covariance = self.kalman_filter.update(self.mean, self.covariance,self.convert_coords(new_tlwh))self.state = State.Trackedself.is_activated = Trueself.score = new_track.scoreself.cls = new_track.clsself.idx = new_track.idxdef convert_coords(self, tlwh):return self.tlwh_to_xyah(tlwh)def mark_lost(self):self.state = State.Lostdef mark_removed(self):self.state = State.Removed@propertydef end_frame(self):return self.frame_id@staticmethoddef next_id():Track.count += 1return Track.count@propertydef tlwh(self):"""Get current position in bounding box format `(top left x, top left y,width, height)`."""if self.mean is None:return self._tlwh.copy()ret = self.mean[:4].copy()ret[2] *= ret[3]ret[:2] -= ret[2:] / 2return ret@propertydef tlbr(self):"""Convert bounding box to format `(min x, min y, max x, max y)`, i.e.,`(top left, bottom right)`."""ret = self.tlwh.copy()ret[2:] += ret[:2]return ret@staticmethoddef reset_id():Track.count = 0@staticmethoddef tlwh_to_xyah(tlwh):"""Convert bounding box to format `(center x, center y, aspect ratio,height)`, where the aspect ratio is `width / height`."""ret = numpy.asarray(tlwh).copy()ret[:2] += ret[2:] / 2ret[2] /= ret[3]return ret@staticmethoddef tlbr_to_tlwh(tlbr):ret = numpy.asarray(tlbr).copy()ret[2:] -= ret[:2]return ret@staticmethoddef tlwh_to_tlbr(tlwh):ret = numpy.asarray(tlwh).copy()ret[2:] += ret[:2]return retdef __repr__(self):return f'OT_{self.track_id}_({self.start_frame}-{self.end_frame})'class BYTETracker:def __init__(self, frame_rate=30):self.tracked_tracks = []self.lost_tracks = []self.removed_tracks = []self.frame_id = 0self.max_time_lost = int(frame_rate)self.kalman_filter = KalmanFilterXYAH()self.reset_id()def update(self, boxes, scores, object_classes):self.frame_id += 1activated_tracks = []re_find_tracks = []lost_tracks = []removed_tracks = []# add indexboxes = numpy.concatenate([boxes, numpy.arange(len(boxes)).reshape(-1, 1)], axis=-1)indices_low = scores > 0.1indices_high = scores < 0.5indices_remain = scores > 0.5indices_second = numpy.logical_and(indices_low, indices_high)boxes_second = boxes[indices_second]boxes = boxes[indices_remain]scores_keep = scores[indices_remain]scores_second = scores[indices_second]cls_keep = object_classes[indices_remain]cls_second = object_classes[indices_second]detections = self.init_track(boxes, scores_keep, cls_keep)""" Add newly detected tracklets to tracked_stracks"""unconfirmed = []tracked_stracks = []for track in self.tracked_tracks:if not track.is_activated:unconfirmed.append(track)else:tracked_stracks.append(track)""" Step 2: First association, with high score detection boxes"""track_pool = self.joint_stracks(tracked_stracks, self.lost_tracks)# Predict the current location with KFself.multi_predict(track_pool)#print('ddd')dists = self.get_dists(track_pool, detections)matches, u_track, u_detection = linear_assignment(dists, thresh=0.8)for tracked_i, box_i in matches:track = track_pool[tracked_i]det = detections[box_i]if track.state == State.Tracked:track.update(det, self.frame_id)activated_tracks.append(track)else:track.re_activate(det, self.frame_id, new_id=False)re_find_tracks.append(track)""" Step 3: Second association, with low score detection boxes"""# association the untrack to the low score detectionsdetections_second = self.init_track(boxes_second, scores_second, cls_second)r_tracked_tracks = [track_pool[i] for i in u_track if track_pool[i].state == State.Tracked]dists = iou_distance(r_tracked_tracks, detections_second)matches, u_track, u_detection_second = linear_assignment(dists, thresh=0.5)for tracked_i, box_i in matches:track = r_tracked_tracks[tracked_i]det = detections_second[box_i]if track.state == State.Tracked:track.update(det, self.frame_id)activated_tracks.append(track)else:track.re_activate(det, self.frame_id, new_id=False)re_find_tracks.append(track)for it in u_track:track = r_tracked_tracks[it]if track.state != State.Lost:track.mark_lost()lost_tracks.append(track)"""Deal with unconfirmed tracks, usually tracks with only one beginning frame"""detections = [detections[i] for i in u_detection]dists = self.get_dists(unconfirmed, detections)matches, u_unconfirmed, u_detection = linear_assignment(dists, thresh=0.7)for tracked_i, box_i in matches:unconfirmed[tracked_i].update(detections[box_i], self.frame_id)activated_tracks.append(unconfirmed[tracked_i])for it in u_unconfirmed:track = unconfirmed[it]track.mark_removed()removed_tracks.append(track)""" Step 4: Init new stracks"""for new_i in u_detection:track = detections[new_i]if track.score < 0.6:continuetrack.activate(self.kalman_filter, self.frame_id)activated_tracks.append(track)""" Step 5: Update state"""for track in self.lost_tracks:if self.frame_id - track.end_frame > self.max_time_lost:track.mark_removed()removed_tracks.append(track)#print('ccc')self.tracked_tracks = [t for t in self.tracked_tracks if t.state == State.Tracked]self.tracked_tracks = self.joint_stracks(self.tracked_tracks, activated_tracks)self.tracked_tracks = self.joint_stracks(self.tracked_tracks, re_find_tracks)self.lost_tracks = self.sub_stracks(self.lost_tracks, self.tracked_tracks)self.lost_tracks.extend(lost_tracks)self.lost_tracks = self.sub_stracks(self.lost_tracks, self.removed_tracks)self.removed_tracks.extend(removed_tracks)self.tracked_tracks, self.lost_tracks = self.remove_duplicate_stracks(self.tracked_tracks, self.lost_tracks)output = [track.tlbr.tolist() + [track.track_id,track.score,track.cls,track.idx] for track in self.tracked_tracks if track.is_activated]return numpy.asarray(output, dtype=numpy.float32)@staticmethoddef init_track(boxes, scores, cls):return [Track(box, s, c) for (box, s, c) in zip(boxes, scores, cls)] if len(boxes) else []  # detections@staticmethoddef get_dists(tracks, detections):dists = iou_distance(tracks, detections)dists = fuse_score(dists, detections)return dists@staticmethoddef multi_predict(tracks):Track.multi_predict(tracks)@staticmethoddef reset_id():Track.reset_id()@staticmethoddef joint_stracks(tlista, tlistb):exists = {}res = []for t in tlista:exists[t.track_id] = 1res.append(t)for t in tlistb:tid = t.track_idif not exists.get(tid, 0):exists[tid] = 1res.append(t)return res@staticmethoddef sub_stracks(tlista, tlistb):stracks = {t.track_id: t for t in tlista}for t in tlistb:tid = t.track_idif stracks.get(tid, 0):del stracks[tid]return list(stracks.values())@staticmethoddef remove_duplicate_stracks(stracksa, stracksb):pdist = iou_distance(stracksa, stracksb)pairs = numpy.where(pdist < 0.15)dupa, dupb = [], []for p, q in zip(*pairs):timep = stracksa[p].frame_id - stracksa[p].start_frametimeq = stracksb[q].frame_id - stracksb[q].start_frameif timep > timeq:dupb.append(q)else:dupa.append(p)resa = [t for i, t in enumerate(stracksa) if i not in dupa]resb = [t for i, t in enumerate(stracksb) if i not in dupb]return resa, resb

本文来自互联网用户投稿，该文观点仅代表作者本人，不代表本站立场。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如若转载，请注明出处：http://www.mzph.cn/news/96591.shtml

如若内容造成侵权/违法违规/事实不符，请联系多彩编程网进行投诉反馈email:809451989@qq.com，一经查实，立即删除！

相关文章

MDC方式实现简单链路追踪

MDC方式实现简单链路追踪

MDC 方式实现日志链路追踪 拦截器 package com.cdn.log.interceptor;import com.cdn.log.consts.CLogConst; import com.cdn.log.utils.IdUtil; import org.slf4j.MDC; import org.springframework.util.StringUtils; import org.springframework.web.servlet.ModelAndView; im…
阅读更多...
山西电力市场日前价格预测【2023-10-08】

山西电力市场日前价格预测【2023-10-08】

日前价格预测 预测说明&#xff1a; 如上图所示&#xff0c;预测明日&#xff08;2023-10-08&#xff09;山西电力市场全天平均日前电价为258.40元/MWh。其中&#xff0c;最高日前电价为496.19元/MWh&#xff0c;预计出现在18:45。最低日前电价为0.00元/MWh&#xff0c;预计出…
阅读更多...
PMP考试需要多少钱?费用总计来啦！

PMP考试需要多少钱?费用总计来啦！

PMP项目管理专业人士资格认证是由项目管理协会&#xff08;Project Management Institute&#xff0c;简称PMI&#xff09;发起的。PMP作为世界级的项目管理认证证书&#xff0c;拥有着先进的项目管理知识体系&#xff0c;它严格评估项目管理人员知识技能是否具有高品质的资格认…
阅读更多...
【FreeRTOS】内存管理简单介绍

【FreeRTOS】内存管理简单介绍

有没有想过什么移植FreeRTOS时&#xff0c;为什么有多种的内存文件&#xff0c;我们工程只使用Heap_4&#xff0c;其他的有什么用&#xff1f;每个的区别是什么&#xff1f;FreeRTOS是一种流行的实时操作系统&#xff0c;广泛应用于嵌入式系统开发中。在嵌入式系统中&#xff0…
阅读更多...
JEPG Encoder IP verilog设计及实现

JEPG Encoder IP verilog设计及实现

总体介绍： 采用通用的常规 Verilog 代码编写，可用于任何 FPGA。 该内核不依赖任何专有 IP 内核，而是用 Verilog 编写了实现 JPEG 编码器所需的所有功能，代码完全独立。 编码器内核的输入是一条 24 位数据总线，红色像素、绿色像素和蓝色像素各有 8 位。 信号 "data_i…
阅读更多...
记录使用vue-test-utils + jest 在uniapp中进行单元测试

记录使用vue-test-utils + jest 在uniapp中进行单元测试

目录 前情安装依赖package.json配置jest配置测试文件目录编写setup.js编写第一个测试文件jest.fn()和jest.spyOn()jest 解析scss失败测试vuex$refs定时器测试函数调用n次手动调用生命周期处理其他模块导入的函数测试插槽 前情 uniapp推荐了测试方案dcloudio/uni-automator&…
阅读更多...
vue+vite项目静态文件引用丢失

vue+vite项目静态文件引用丢失

排查问题&#xff1a; 查看vite.config.js中是否配置 base: ./, 如果配置了&#xff0c;那么 本地运行 例如&#xff1a;css中引入字体&#xff0c;可将字体文件直接放在public文件夹下通过"/"引入 如果需要动态引入图片&#xff0c;可通过 import defaultA fr…
阅读更多...
（三）行为模式：8、状态模式（State Pattern）（C++示例）

（三）行为模式：8、状态模式（State Pattern）（C++示例）

目录 1、状态模式&#xff08;State Pattern&#xff09;含义 2、状态模式的UML图学习 3、状态模式的应用场景 4、状态模式的优缺点 &#xff08;1&#xff09;优点 &#xff08;2&#xff09;缺点 5、C实现状态模式的实例 1、状态模式&#xff08;State Pattern&#x…
阅读更多...
光伏发电预测（LSTM、CNN_LSTM和XGBoost回归模型，Python代码）

光伏发电预测（LSTM、CNN_LSTM和XGBoost回归模型，Python代码）

运行效果&#xff1a;光伏发电预测&#xff08;LSTM、CNN_LSTM和XGBoost回归模型&#xff0c;Python代码&#xff09;_哔哩哔哩_bilibili 运行环境库的版本 光伏太阳能电池通过互连形成光伏模块&#xff0c;以捕捉太阳光并将太阳能转化为电能。因此&#xff0c;当光伏模块暴露…
阅读更多...
【面试高高手】—— Java垃圾回收

【面试高高手】—— Java垃圾回收

文章目录 1.什么是Java垃圾回收&#xff08;Garbage Collection&#xff09;&#xff1f;为什么需要它&#xff1f;2. Java中的垃圾回收是如何工作的&#xff1f;请简要解释垃圾回收的工作原理。3.什么是垃圾对象&#xff08;Garbage Objects&#xff09;&#xff1f;如何确定一…
阅读更多...
深入探究 C++ 编程中的资源泄漏问题

深入探究 C++ 编程中的资源泄漏问题

目录 1、GDI对象泄漏 1.1、何为GDI资源泄漏&#xff1f; 1.2、使用GDIView工具排查GDI对象泄漏 1.3、有时可能需要结合其他方法去排查 1.4、如何保证没有GDI对象泄漏&#xff1f; 2、进程句柄泄漏 2.1、何为进程句柄泄漏&#xff1f; 2.2、创建线程时的线程句柄泄漏 …
阅读更多...
成功解决@Async注解不生效的问题，异步任务处理问题

成功解决@Async注解不生效的问题，异步任务处理问题

首先&#xff0c;有这样一个异步监听方法 然后配置好了异步线程池 package com.fdw.study.config;import org.springframework.context.annotation.Bean; import org.springframework.context.annotation.ComponentScan; import org.springframework.context.annotation.Conf…
阅读更多...
国产1.8V低电压输入，可用于驱动步进电机；H 桥驱动电路单元可以直接驱动IR-CUT

国产1.8V低电压输入，可用于驱动步进电机；H 桥驱动电路单元可以直接驱动IR-CUT

D6212是专为安防摄像头系统设计的驱动电路&#xff0c;电路由八路达林顿管 阵列和 H 桥驱动电路两个单元组成。八路达林顿管阵列均带有续流二极 管&#xff0c;可用于驱动步进电机&#xff1b;H 桥驱动电路单元可以直接驱动IR-CUT。单个 达林顿管在输入电压低至 1.8V 状态下支持…
阅读更多...
口袋参谋：如何提升宝贝流量？这三种方法超实用！

口袋参谋：如何提升宝贝流量？这三种方法超实用！

​你的店铺能不能出爆款&#xff1f;提升单品流量是关键。 对于新手卖家来说&#xff0c;是缺乏运营技巧和运营经验的&#xff0c;运营技巧主要体现在标题写作、各种图片和视频制作等。 由于新手买家没有经验&#xff0c;习惯于直接使用数据包上传&#xff0c;导致宝贝没有展…
阅读更多...
Java卷上天，可以转行干什么？

Java卷上天，可以转行干什么？

小刚是某名企里的一位有5年经验的高级Java开发工程师&#xff0c;每天沉重的的工作让他疲惫不堪&#xff0c;让他萌生出想换工作的心理&#xff0c;但是转行其他工作他又不清楚该找什么样的工作 因为JAVA 这几年的更新实在是太太太……快了&#xff0c;JAVA 8 都还没用多久&am…
阅读更多...
Head-First-Design-Patterns

Head-First-Design-Patterns

目录 1.策略模式 2.观察者模式 3.装饰者模式 4.工厂模式 5.单件模式 6.命令模式 7.适配器和外观模式 8.模板方法模式 9.迭代器和组合模式 10.状态模式 11.代理模式 12.复合模式 13.更好与设计模式相处 代码地址&#xff1a;https://github.com/wangzs2018/Design-…
阅读更多...
cpp primer plus笔记01-注意事项

cpp primer plus笔记01-注意事项

cpp尽量以int main()写函数头而不是以main()或者int main(void)或者void main()写。 cpp尽量上图用第4行的注释而不是用第5行注释。 尽量不要引用命名空间比如:using namespace std; 函数体内引用的命名空间会随着函数生命周期结束而失效&#xff0c;放置在全局引用的命名空…
阅读更多...
22. 括号生成

22. 括号生成

22. 括号生成 回溯 class Solution {List<String> res new ArrayList<>();StringBuilder path new StringBuilder(); // path推荐这种写法public List<String> generateParenthesis(int n) {backtrack(n, 0, 0);return res;}void backtrack(int n, int …
阅读更多...
PX4仿真添加world模型文件，并使用yolov8进行跟踪

PX4仿真添加world模型文件，并使用yolov8进行跟踪

前言 目的：我们是为了在无人机仿真中使用一个汽车模型，然后让仿真的无人机能够识别到这个汽车模型。所以我们需要在无人机仿真的环境中添加汽车模型。 无人机仿真中我们默认使用的empty.world文件，所以只需要将我们需要的模型添加到一起写进这个empty.world文件中去就可以…
阅读更多...
通过套接字手动写一个回显服务器吧

通过套接字手动写一个回显服务器吧

背景：程序员主要编写应用层的代码。真正要发送的数据需要上层协议调用下层协议，而应用层调用传输层时，传输层(系统内核)给应用层提供的一组API统称为Socket API。 系统提供给Java程序员的Socket API主要有两组： 基于UDP的API基于TCP的API目录 一、为什么需要网络编程？——…
阅读更多...
最新文章

Copyright @ 2022~2023